Structural Evolution and Pharmacology of a Novel Series of Triacid

J.Med. Chem. 1994,37,4508-4521

4608

Structural Evolution and Pharmacology of a Novel Series of Triacid Angiotensin I1 Receptor Antagonists Alan D. Palkowitz,* Mitchell I. Steinberg, K. Jeff Thrasher, Jon K. Reel, Kenneth L. Hauser, Karen M. Zimmerman, Sally A. Wiest, Celia A. Whitesitt, Richard L. Simon, William Heifer, Sheryl L. Lifer, Donald B. Boyd, Charles J. Barnett, Thomas M. Wilson, Jack B. Deeter, Kumiko Takeuchi, Robert E. Riley, William D. Miller, and Winston S. Marshall Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46285 Received May 6, 1994@

cis-4-(4-Phenoxy)-l-[l-oxo-2(R)-[4-[(2-sulfobenzoyl)amino)-~-imidazol-l-ylloctyll-~-proline derivatives represent a novel class of potent nonpeptide angiotensin I1 (Ang 11) receptor antagonists. These compounds evolved from directed structure-activity relationship ( S A R ) studies on a lead identified by random screening. Further S A R studies revealed that acidic modification of the 4-phenoxy ring system produced a series of triacid derivatives possessing oral activity in pithed rats. The most potent compound, cis-4-[4-(phosphonomethyl)phenoxyl1-[1-oxo-2(R)-[4-[(2-sulfobenzoyl)amino]-l~-imidazol-l-yl]octyl]-~-proline (le), inhibited the pressor response to exogenously administered Ang I1 for periods up to 8 h following oral dosing. The antihypertensive activity of l e was evaluated in the Lasix-pretreated conscious spontaneously hypertensive rat (SHR)where it produced a dose-dependent fall in blood pressure following oral dosing lasting > 12 h. Antagonists such as le may serve as useful therapeutic agents for the treatment of hypertension as well as for studying the role of Ang I1 in various disease states.

Introduction Recent clinical and experimental studies have provided mounting evidence that angiotensin (Ang 11)converting enzyme inhibitors are effective therapy in a wide variety of cardiovascular indications such as heart failure, myocardial hypertrophy, and diabetic nephro~ a t h y . l In - ~view of the diverse pathophysiologic actions of Ang 11, interfering with the function of the reninangiotensin system through other means continues to be an attractive target for new drug devel~pment.~,~ The recent introduction of potent, orally active receptor antagonists of Ang I1 has added additional impetus t o this efforte7 Thus, there are now a wide variety of antagonists described, most deriving from and/or containing the biphenyl-tetrazole substructure present in losartan (Chart 1).8-11 Recently, we described the synthesis and in vitro pharmacological evaluation of a novel series of diastereomeric phenoxyproline octanoamides as Ang I1 (ATd receptor antagonists.12 Studies comparing functional inhibition of AT1 receptor-mediated responses showed that compound lb (Chart 11, possessing the (R,S,S) configuration, to be significantly more potent (10-1000fold) than any of its seven stereoisomers. Compound lb evolved from a series of substituted imidazole hexanoic acids (see Chart 2) identified by a random screening effort t o uncover potent nonpeptide Ang I1 receptor antag0ni~ts.l~ Although the initial leads from this effort were relatively weak as Ang I1 receptor antagonists in isolated tissues (Chart 21, they led to the discovery of a potent series of phenoxyproline triacids, exemplified by compounds la-g. In this paper, we report the structural evolution of this novel class of Ang I1 antagonists, as well as the in vivo pharmacology of these agents.

* Please send all correspondence to: Dr. Alan D. Palkowitz, Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, IN 46285. Abstract published in Advance ACS Abstracts, November 15,1994. @

Chemistry In order to prepare the compounds shown in Tables 1and 2, we sought a flexible approach that would allow us to readily vary the substitution of the aryloxy ring as well as provide easy access to the desired (R,S,S) diastereomer. For this discussion, the synthetic methods used for the preparation of la-g will be described in detail as they are representative of the methods used to explore the S A R of the entire series.14 Thus, the synthesis of la-g began with the Mitsunobu coupling of phenol derivatives 2a-g with (2S,4R)-N-Cbz-4-hydroxyproline methyl ester (DEAD, Ph3P, THF) to give (2S,4S)-N-Cbz-4-phenoxyproline derivatives 3a-g in 3 7 4 3 % yield.15J6Phenol substrates 2a-c were prepared by esterification of the commercially available carboxylic acids (MeOH or EtOH, pTsOH, reflux). Tetrazole derivative 2g was prepared from (Chydroxyphenyl)acetonitrile, and 2d-f were prepared as described in Scheme l.17-19Removal of the Cbz protecting groups was accomplished by catalytic hydrogenation (EtOH-EtOAc, 10% Pd/C, 40 psi) to give the requisite intermediate 4-phenoxyproline ester derivatives 4a-g. The carboxylic acid derivatives (4a-d), and the tetrazole 4g, were converted to the HBr salt form (ethereal-HBr) for characterization. Construction of the key stereochemically defined 4-nitroimidazole proline octanoamide intermediates 6a-e,g from 4a-e,g is shown in Scheme 2. 4-Nitroimidazole was converted to the octanoic acid derivative (&)-5 by alkylation with ethyl 2-bromooctanoate followed by ester hydrolysis in 90% yield. Nitroimidazole acid (&I-5was converted to its acid chloride ((COC112, CHzClz) and reacted with proline ester derivatives 4ae. In the cases where the HBr salt was employed, the free base was generated by treatment with diisopropylamine prior to reaction with the acid chloride. In the coupling reaction, diastereomers 6a-e and 7a-e were produced as 1:l mixtures and readily separated by flash chromatography.

0022-2623i94i~a3~-450a~a4.5aia0 1994 American Chemical Society

Triacid Angiotensin ZZ Antagonists

Journal

"s,

Chart 1

of

Medicinal Chemistry, 1994, Vol. 37, No. 26 4509

*

0

-p

0

CI

Losartan

COzH

IC R s Id R = l e RI 1f R =

lg R =

Chart 2

40 compounds

290 compounds I

I d

C

O

,

"

d

C

O

z

dC02H

H

11

12

13

b=71.5 f 18.8 pM (n = 19)

KB= 12.5 f 2.5 p M (n I15)

KB= 684.5 k 64.2 nM (n = 1s)

Lead Structurefrom Semen

d

C

O

z

H

14

15

9

KBI 158.6 t 26.8 nM (n= 4)

KB 423.7 f 7.8 nM (n= 60)

KB= 81.7 f 7.2nM (n 25)

Scheme 1 on Cbz' CO2CH3 Cbr'

no

N q S

a

2a R I CO#IzCH3 2b R iCHzC02CHzCH3 2~ R C H Z C H ~ C O Z C H ~ C H ~ 2d R I C(CH&C02CH3 2e R CH2P0(OCH& 2l R I CH2CH2PO(OCH2CH3)2 2g R I CHzCN&Ph3

w

o

n of Id-f

4-mDthoxyphenylaceticacld 4-benzyloxybenzylchlorldo

4-benzyloxybanzaldehyde

4e,f g,b

hb

*

2d

~

2e

~

21

a Conditions: (a) DEAD, PhSP, THF, room temperature; (b) Hz (40 psl), 10% Pd/C, EtOH; (c) ethereal HBr, 25 "C (4a-d, g). (d) LDA, MeI; (e) pyridine hydrochloride, 150 "C; (0 MeOH, pTsOH(cat.), reflux; (g) (Me0)2POH, NaH, THF; (h) ((Et0)2PO)&H2, NaH, DMF.

In subsequent work, it was found that the R-enantiomer of 6 could be readily obtained by selective crystallization of the (-)-cinchonidine salt (Scheme 2). The enantiomeric excess of the resolved acid was found to be 299% after a single recrystallization from EtOHH20. Enantiomeric excess was determined by chiral

HPLC analysis of the methyl ester (CHzN2, EtzO). Using the resolution protocol, 6g was prepared stereoselectively by the DCC coupling of 4g to (R)-5. The absolute stereochemical assignment of 6a-e,g based on the X-ray crystalwas established as (R,S,S) lographic analysis of a closely related compound, 7h

4510 Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26

Palkowitz et al.

Table 1. Structure, Physical Properties, and in Vitro Antagonism of Ang I1 by 16-40

compd 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

R

stereochemistry (X,S,S)

mp, "C 190-195 163-170 198-205 180-190 > 200 dec 155-165 175-185 162-170 160-175 dec 155-162 dec 154-165 dec 145-155 148-155 150-162 138-145 170-175 169-175 152-162 170-180 dec 170-190 dec 170-180 dec 185-190 dec 168-172 dec 180-183 dec 225-230 dec

formula" CznHmNaO.iS

analyses C. H. N

in vitro KB f SE (nM? 50.9 f 12.7 (5) 461 (1) 25.8 (1) 0.80 f 0.12 (20) 3.2 f 1.9 (3) 1.09f 0.5 (7) 25.1 (1) 125.9 (1) 3.0 f 1.7 (3) 34.7 (1) 10.0 (1) 0.44 f 0.12 (11) 11.6 (2) >30 (2) 4.1 f 2.9 (3) 5.2 f 1.4 (4) 0.92 f 0.4 (4) 15.8 (1) 4.1 f 1.2 (7) 4043.4 (2) 0.63 f 0.16 ( 6 ) 3.3 f 1.6 (3) 3.6 (2) 148 (1) 94.6 (2)

H H OH OPh OPh OPh(4-Me) OPh(4-iPr) OPh(4-tBu) OPh(4-F) OPh(4-CF3) OPh(4-Ph) OPh(4-OMe) OPh(3-OMe) OPh(2-OMe) OPh(4-OiPr) OPh(4-OtBu) OPh(3,4-OCHz0-) OPh(3,4-di-OMe) O(2-naphthyl) O(1-naphthyl) O(5-benzofuran) O(5-isoquinoline) O(5-thianaphthene) O(3-pyridyl) O(5-isoxazole) 40 All compounds had C, H, and N microanalysis within f0.4% theoretical value unless otherwise noted. N: calcd, 8.50; found, 7.73. N calcd, 8.90; found, 8.42. N: calcd, 9.90; found, 10.49. e Numbers in parentheses represent the number of individual experiments, with each experiment representing the average of at least four tissues. Table 2. Physical Properties and in Vitro and in Vivo Antagonism of Ang I1 by la-g and 41-42

compd

mp, "C formula" analyses C, H, N la 185-195 C3oH34N40ioS c,H, N lb 160-175 C31H3~N40ioS C, H, N IC 140-148 C3zH3sN40ioS Id 182-187 C,CH,~N C~~H~ON~O~OS@.~H C, H, N le 190 dec C3oH3dWW" If 230-234 C, H, N C3iH3gN4011PS*1.5HCl C, H, N 1g 165-172 C~~H~~NBOBS C, H, N 41 166-171 C3iH36N40ioS C, H, N 42 153-167 C3iH36N40ioS All compounds had C, H, and N microanalysis within &0.4% theoretical value. individual experiments. NA = not active (no shiR in the dose response curve noted

(Figure 1, precursor to 20 (Table 1)). This compound differs only in the lack of substitution at the para position of the 4-(aryloxy)proline group (i.e., R = H). Isomers 6h and 7h possess lH NMR spectral characteristics and silica gel mobility profiles that are directly analogous to those observed for the para-substituted derivatives 6a-e,g and 7a-e, as well as the intermediates leading to 21-40.20 The 'H NMR spectra of the (R,S,S) isomers in DMSO-& produce two sets of singlets (doubling due to amide rotamers) at 8.45 and 8.43 ppm, and at 7.99 and 7.95 ppm, respectively, that are the resonances of the two imidazole protons. For the (S,S,S) isomers, the analogous signals are observed at 8.43 and 8.30 ppm and at 7.94 and 7.78 ppm. These spectral characteristics are independent of the substitution on the proline phenoxy ring and are remarkably consistent

in vitro in vivo f SE ( d V f ) b Kb f SE (mgkg, po)bzc 1.1 f 0.3 (5) NA (3) 0.27 f 0.05 (17) 2.8 f 0.2 (16) 0.6 f 0.3 (3) NA (3) 6.5 f 0.3 (3) 0.6 f 0.3 (3) 1.8 f 0.4 (16) 0.9 f 0.5 (8) 0.4 f 0.4 (3) 8.3 f 1.0 (4) 0.6 f 0.5 (3) NA (3) 158, 501 (2) NA (2) 10.6 f 5.0 (4) 6.3, 15.8 (2) Numbers in parentheses represent the number of at this dose and time, i.e. '10 mgkg). Kb

from compound to compound, differing only within f 0.02 ppm. Consequently, the lH NMR spectra provided a simple means of confirming stereochemical assignment as well as assessing diastereomeric purity of the isomers. Additionally, in all solvent systems investigated, the (R,S,S)isomers had greater Rfvalues on silica gel than the corresponding (S,S,S)isomers. The elaboration of 6a-e,g to la-e,g is shown in Scheme 3. Catalytic reduction of Ga-e,g (EtOAc-EtOH, 10% PdK, 40 psi) generated the 4-aminoimidazole intermediate. In our hands, it was not possible to isolate or store the 4-aminoimidazole due to problems of stability. Hence, this material was reacted immediately with sulfobenzoicanhydride (THF or CH&N, 25 "C). After 1 h, trituration of the crude reaction mixture provided sulfonic acid derivatives 8a-e,g in

Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26 4511

Triacid Angiotensin II Antagonists

Scheme 2

02N

f A

C

O

z

H

(W-6

Conditions: (a) NaH, ethyl 2-bromooctanoate,DMF, 25 "C; (b) NaOH, THF; (c) (COCl)z, CHzClz, DMF(cat.);(d) 4a-e, TEA, CHzClz; (e) (-)-cinchonidine, EtOH-H20; (0 DCC, HOBT, 4g, DMF. a

7h (S,S,S) 6h I(R,S,S) (noishown)

Figure 1. ORTEP drawing of the crystal structure of 7h (S,S,S).

Scheme 3

0. R 3 COzCHnCH3 6b R ICHzCOzCHzCH3 8c R = CHzCHZC02CHzCH3

W R I C(CH&COzCH3 80 R I CHzPO(OCH3)z 6g R D CHZCN~H a

I). R I COzCHzCHs

8b R I CHzCOZCHzCH3 CHzCHzCOzCHzCH3 W R I C(CH3hCOzCH3 I). R I CHzPO(OCH3h 8g R I CH2CN.H OC R

Conditions: (a) Hz (40 psl), 10%PaC, EtOH; (b) sulfobenzoic anhydride, THF; (c) NaOH (Sa-d,g); (d) TMS-Br,CHzClz then NaOH

@e).

good yield. In general, it was not necessary to rigorously purify the esters prior to hydrolysis. The conversion of 8a-d,g to la-d,g was accomplished by treatment with 1N NaOH, followed by acidification to pH = 1.5 with 5 N HC1. The triacids were isolated either by direct filtration of the aqueous solution or by extraction into EtOAc-EtOH, followed by concentration and trituration

of the crude product from CH&N/Et20. For 8e, TMSBr cleavage of the dimethyl phosphonate in CH2C12 at 0 "C followed by basic workup provided l e as a white solid after acidification and isolation using the extraction-trituration protocol.21 For the preparation of ethylenephosphonic acid lf,we took advantage of the resolution chemistry described

4612 Journal of Medicinal Chemistry, 1994, Vol. 37,No. 26

Palkowitz et al.

Scheme 4

9 a Conditions: MeOH, pTsOH(cat.), reflux; (b) H2 (40 psl), 10%P d C , EtOH; (c) sulfobenzoic anhydride, THF; (d) NaOH; (e) DCC, HOBT, 4g, DMF; (0 TMS-Br, CHzClz then NaOH.

earlier. As detailed in Scheme 4, esterification of (R)-6 followed by catalytic reduction, acylation with sulfobenzoic anhydride, and careful alkaline hydrolysis provided enantiomerically enriched acid 9 in 36% overall yield. Although no attempts were made to determine the enantiomeric purity of this material, we demonstrated that esterification of (R1-5(MeOH, cat. pTsOH) followed by hydrolysis with 2.0 equiv of NaOH resulted in no measurable loss of ee. Hence, DCC coupling, of 9 to 4f gave diester Sf,isolated in 30% yield by flash chromatography (SiOz, 10% MeOWCHC13). lH NMR analysis showed this material to be a single diastereomer. Finally, treatment with TMS-Br in CHzClz followed by hydrolytic workup (NaOH) provided If in 26% yield. The low yield is representative of the difficulty encountered in isolating these polar species from the aqueous media. However, no exhaustive attempts were made to optimize isolation procedures.

Results and Discussion Our efforts toward the discovery of nonpeptide Ang

I1 antagonists grew out of a large volume receptor-based screening assay using the rat adrenal glomerulosa preparation.22 After examining numerous compounds, imidazolehexanoic acid 11 emerged as a lead structure from which extensive SAFt studies were begun. "his compound proved to be a weak, competitive antagonist of Ang I1 in isolated rabbit aorta, possessing a KBof 71.5 pM. As shown in Chart 2, early structural modifications revealed that reversal of the amide linkage from the imidazole to the benzoyl ring and extension of the aliphatic side chain from four to six carbons led to a slight increase in activity (12). More significant increases in activity were not realized until acidic substitutions were introduced at the ortho position of the benzoyl ring (13).Examination of acid isosteres showed that the sulfonic acid derivative 9 was the most potent with a KB of 81.7 nM.23The discovery of 9 coincided with the early publications of the losartan structure and S A R by the DuPont group. It was apparent to us that 9 shared some structural and SAR similarity to the DuPont series, especially with regard to overall dimension, the need for acidic modification of the benzoyl group at the ortho position, and the requirement of a lipophilic chain. However, our compounds were significantly less potent in vitro than either losartan itself or its carboxylic acid metabolite EXP3174.24 For comparison, losartan yielded a KB of 6.3 nM under the same

condition^.^^ In exploration of the S A R centered around the carboxylic acid of 9, we found that substitution with L-proline produced a set of diastereomeric derivatives (16 and 17,Table 1)with markedly different in vitro

activity. The (R,S)isomer 16 gave a KB of 50.9 nM, while the (S,S) isomer 17 gave a KB of 461 nM. This result indicated that there is a definite stereochemical preference of the hexyl side chain for interaction with the AT1 receptor. Additionally, the modest increase in activity gained by introduction of the proline into 9 revealed that repositioning of the carboxyl group as part of the proline system was beneficial to activity. Hence, the proline ring system of 16 provided an attractive template from which to expand our SAR studies. Due to the availability of ~-(2S,4R)-4-hydroxyproline, we chose to explore modification of the proline ring by chemical transformations derived from the 4-hydroxyl group. While the cis-4-hydroxyprolinederivative 18 was weakly active, a large increase in in vitro potency (150fold) was achieved with the introduction of a 4-phenoxy group cis to the carboxylic acid (19,Table 1). Additionally, while all of our previous compounds were competitive antagonists of Ang I1 in the rabbit aorta, the more potent phenoxyproline derivative 19 was a noncompetitive antagonist. This is consistent with other studies that demonstrate nonsurmountable antagonism of Ang I1 with potent diacidic compounds, such as CV-11194, EXP3174, and GR 117289.11,24926 As was observed for the simple proline case, there was a large separation in activity between the (R,S,S) and (S,S,S) isomers.27 Consequently, all subsequent S A R studies were conducted with derivatives possessing the (R,S,S)absolute stereochemistry. The large increase in in vitro potency achieved by the introduction of the phenoxyproline side chain into 9 elevated this series to a comparable level of potency with other known nonpeptide Ang I1 antagonists. To more fully characterize the interaction of the phenoxy group with the receptor, we investigated structural modifications of the aryl ring as summarized in Table 1. It was possible to substitute the aryl ring at the para position with small alkyl groups and alkyl ethers and maintain good potency (21 and 27). However, activity dropped off sharply when large sterically demanding groups were introduced (22,231.Additionally, there appeared t o be a regiochemical preference for para substitution; the ortho and meta methoxy derivatives 29 and 28 were significantly less potent. Poly acyclic substitution was not well-tolerated as exemplified by the 3,4-dimethoxy derivative 33;however, the less sterically demanding 3,4-methelenedioxy compound 32 retains excellent potency. Finally, activity was reduced by direct para substitution with strongly electron withdrawing groups such as F and CF3 (24and 25,respectively). In addition t o substituted aryl derivatives, we explored various bicyclic aromatic systems, also shown in

Triacid Angiotensin 11Antagonists

Journal

of

Medicinal Chemistry, 1994, Vol. 37, No. 26 4513

Chart 3 1

140

1

-S03H

E

5

iw V.hlCIe 0 2 nr (4) 4 Ht (10) 8 nr (3) 18Hr(3)

41 .Ol

1

1

io

100

Ang I1 (rgiksJ.v.)

Figure 2. Time course of le antagonism of exogenous Ang I1 in pithed rats. Compound l e (10 mgkg) or vehicle (0.1 mL of 0.1 N NaOH diluted to -2 mL with distilled HzO) was administered to normotensive Sprague-Dawley rats by gavage. At the times indicated, rats were pithed and the mean pressor response to various iv doses of Ang I1 determined. Values are the mean & SE of the number of experiments indicated in parentheses. MAP = mean arterial pressure. 42

Table 1. The 5-substituted benzofuran derivative 36 was equipotent with 19; however, activity was reduced with the more sterically demanding 2-naphthalene (34), thianaphthene (38), and isoquinoline (37). Interestingly, activity was markedly decreased with the l-naphthy1 derivative 35. This result was consistent with the reduction in potency observed with ortho and meta substitution of the aryloxy ring as discussed earlier. As a final point, attempts to replace the phenoxy group with other oxygen linked heterocycles such as 3-pyridyl 39 and 5-isoxazole 40 resulted in a substantial loss of potency. Despite the tremendous increase in i n vitro potency achieved with the introduction of the phenoxyproline moiety, the more potent compounds in Table 1showed no activity against the pressor response to Ang I1 in pithed rats following oral administration at doses up to 30 mgikg.28 Modification of the phenoxyproline series to produce analogs with a long duration of action following oral administration in the pithed rat model was subsequently achieved by appending an acidic function at the para position of the aryl ring. This yielded a series of triacid derivatives la-g (Chart 11, the i n vitro and i n vivo data for which are summarized in Table 2. All compounds showed high potency i n vitro as nonsurmountable Ang I1 antagonists, with KB'S ranging from 0.3 to 1.1 nM calculated using noncompetitive kinetic models (see Methods). Also presented in Table 2 are the ortho and meta acetic acid derivatives 41 and 42,respectively (Chart 3). The poor in vitro activity for these compounds is consistent with a regiochemical preference for substitution of the aryloxy ring at the para position. All compounds in Table 2 were studied for oral activity as antagonists of the pressor response to Ang I1 in pithed rats. To compare potency of the compounds, we determined an i n vivo KB using noncompetitive modeling techniques.28 Of all the compounds in Table 2 tested at an oral dose of 10 mgkg, only lb, ld-f, and 42 showed a measurable shift in the dose response curve to exogenously administered Ang I1 4 h postdosing. This observation is intriguing in that only minor structural differences distinguish these compounds. A requirement for good oral activity in the carboxylic and phosphonic acid series appears t o be a methylene spacer

between the acidic moiety and aryl ring ( l b and le). The homologous phosphonoethyl compound If was less potent than l e , and the corresponding carboxylic acid IC was inactive following oral administration. Additionally, geminal substitution of the methylene group (la)resulted in diminished oral activity. Based on the acid isosteres we investigated, good oral activity was particular to the carboxylic and phosphonic acid derivatives, as the methylenetetrazole analog lg was devoid of oral activity. Of the orally active compounds, phosphonic acid l e was the most potent, with an i n vivo KB of 1.8 mgikg. A time course for inhibition of the pressor response to Ang I1 for le (Figure 2) demonstrated significant i n vivo antagonism for at least 8 h following a single oral dose of 10 mgkg. The inhibition was nonsunnountable i n vivo as anticipated from the in vitro data (Table 2). The improved oral activity observed upon introduction of a third acid function to the phenoxyproline derivative 19 is intriguing. Attempts to replace either of the carboxyl groups of l b with simple alkyl esters resulted in a loss of oral activity when evaluated under the same conditions described for lag. The potential of l e as an orally active antihypertensive agent was evaluated further by studying the compound in Lasix-pretreated (10 mgkg, sc) conscious spontaneously hypertensive rats (SHR). In this model, l e was dosed orally, and mean blood pressure and heart rate were determined by an implanted blood pressure transducer and transmitted by radiotelemetry. Figure 3 shows a dose response relationship with a fall in blood pressure seen at doses as low as 1 mgkg and a maximal effect achieved at 15 mgkg (30 mmHg reduction from baseline). The antihypertensive effect had a rapid onset (within 1h following dosing) and lasted for the duration of the monitoring period ('12 h). Heart rate was generally unchanged; however, at the 1 mgkg dose there was a decrease compared to the vehicle group which reached statistical significance 9 h after dosing (-46 & 13 b/min,p = 0.01) . In a final experiment, we compared the antihypertensive effect of le to other known agents that interfere with the renin-angiotensin pathway. Shown in Figure 4 is the time course of the blood pressure reduction for l e and losartan following a single oral dose of 15 pmoll kg in the non-Lasix pretreated SHR. Also, included in

Palkowitz et al.

4614 Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26

receptor antagonists structurally diverse from the biphenyl-tetrazole-derived agents have been described. For example, SKB 108566, a l-(carboxybenzyllimidazoled-acrylic acid, was designed based on a putative model of the conformation of Ang II.29 Thus,the unique structure of l e ( M w = 694, three ionizable groups at physiologic pH, and three chiral centers) identifies it as a novel entry to the relatively small list of orally effective Ang I1 (AT1) antagonists not derived from losartan. Moreover, the pharmacological profile of l e suggests its utility for studying the role of Ang I1 in many pathologic settings.

Tlme (Hour@

Figure 3. Effect of le on mean pressure of Lasix-pretreated SHR monitored by radio-telemetry. Compound or vehicle was administered by gavage in the doses indicated and mean pressure monitored for 12 h. Mean baseline pressures for the 2 h period before dosing was (mmHg): 147 f 3, 148 f 5, 150 f 5, 143 f 4, and 149 i 5 for the vehicle, 1, 10, 15, and 30 mgkg group, respectively. Mean baseline pressures were not significantly different among groups (ANOVA).In the vehicle group, only the 10 h point was significantly different from baseline (Student's t). In the 1 mgkg group, pressure was significantly decreased from baseline values a t 2, 5, 8, 9, 10, and 11h. All other doses at all time points showed significant decreases from baseline. In general, doses of 10-30 mgkg yielded significant differences from the vehicle group a t most time points. 101

I

+

-30 0

2

4

0 8 Tlma (Hours)

10

12

14

Figure 4. Time course of the effect of losartan (15 pmol/kg, PO),enalapril(55 pmolkg, PO),and le (15pmoYkg, PO) on mean blood pressure in conscious SHR monitored by radiotelemetry. Compound or vehicle was administered by gavage in the doses indicated and mean pressure monitored for 12 h. There were no significant differences in the antihypertensive effects of losartan and l e or enalapril (ANOVA). Significant effects of all three compounds were evident for at least 12 h after dosing.

this study is the angiotensin converting enzyme inhibitor enalapril, given in a larger dose of 55 pmoYkg (PO). All test compounds produced a substantial reduction of blood pressure for '12 h that were not significantly different from one another. In considering this experiment, it is important to note that in rats, losartan is readily transformed into a metabolite (ED31741 that is intrinsically at least 10-fold more potent than le.24

Conclusion In summary, we have identified a structurally novel class of potent nonpeptide Ang I1 (AT11 receptor antagonists with functional in vitro KB values as low as 0.3 nM. Of these compounds, triacids lb, ld-le, and 42 displayed a significant in vivo blockade of the AT1 receptor in pithed rats following oral dosing. The most potent compound of the series, le, produced a dosedependent antihypertensive effect in the Lasix-pretreated conscious SHR with a long duration of action ('12 h) when administered orally. Only a few Ang I1

Experimental Section General. Melting points were determined with a ThomasHoover melting point apparatus and are uncorrected. 'H NMR spectra were obtained on a GE QE-300 spectrometer at 300.15 MHz in the solvent indicated. Field desorption (FD) mass spectra were recorded on a VG Analytical ZAB-3F instrument. High-resolution (HR) and fast atom bombardment ( F B I mass spectra were recorded on a VG Analytical ZABB-SE instrument. Elemental analyses were determined by the Physical Chemistry Department at Lilly Research Laboratories and are within &0.4% of the theoretical values unless otherwise indicated. Methyl 4-Hydro~a,a-dimethylphenylacetate (2d). To a solution of diisopropylamine (20.2 g, 200 mmol) in 200 mL of anhydrous THF at -5 "C under Nz was added n-BuLi (125 mL, 200 mmol, 1.6 M solution in hexanes) dropwise via syringe. &r stirring for 15 min, 4-methoxyphenylacetic acid (8.31 g, 50 mmol) was added in small portions. The mixture was stirred at -5 "C for 30 min and then treated with iodomethane (20.0 mL, 319 mmol). The reaction mixture was allowed t o gradually warm t o room temperature, stirred for 30 min, and then quenched by pouring into 300 mL of saturated NH4Cl solution. The aqueous was extracted with Et20 (3 x 100 mL). The organic was dried (Na~S04)and concentrated in uacuo to give 9.50 g (97%) of 4-methoxy-a,adimethylphenylacetic acid as a white solid: mp 78-81 "C; 'H NMR (DMSO-ds) 6 12.20 (bs, lH), 7.21 (d, J = 8.7 Hz, 2H), 6.84 (d, J = 8.7 Hz, 2H), 3.69 (s, 3H), 1.40 (s, 6H); FD MS 194. Anal. (CllH1403) C, H. 4-Methoxy-a,a-dimethylphenylacetic acid (9.25 g, 47.7 "01) was combined with pyridine hydrochloride (30.0 5. 260 mmol) and heated to 170-190 "C under a NZatmosphere for 5 h. After cooling t o room temperature, the solid residue was partitioned between EtOAc and HzO. The layers were separated, and the organic was extracted several times with HzO. The organic was then dried (NazSO4) and concentrated in uacuo t o give acid as a 8.10 g (94%) of 4-hydroxy-a,a-dimethylphenylacetic tan solid: mp 130-133 "C; 'H NMR (DMSO-ds) 6 12.10 (bs, lH), 9.24 (bs, lH), 7.94 (d, J = 8.6 Hz, 2H), 6.60 (d, J = 8.6 Hz, 2H), 1.37 (s, 6H). Anal. (C10H1203) C, H. 4-Hydroxy-a,a-dimethylphenylacetic acid (8.0 g, 44.4 mmol) was dissolved in 150 mL of anhydrous MeOH along with 3 mL of concentrated HzS04. The mixture was heated to reflux for 12 h. Upon cooling, the MeOH was removed in uacuo. The concentrate was dissolved in Et20 (200 mL) and washed several times with HzO. The organic was then dried (NazS04) and concentrated in uacuo t o provide 8.05 g (93%) of methyl 4-hydroxy-a,a-dimethylphenylacetate(2d) as a white solid: mp 91-94 "C; lH NMR (DMSO-&) 6 9.28 (bs, lH), 7.05 (d, J = 8.5 Hz, 2H), 6.63 (d, J = 8.5 Hz, 2H), 3.52 (8,3H), 1.41 (s, 6H). Anal. (CiiH1403) C, H. Dimethyl (4-Hydroxybenzy1)phosphonate(2e). To a solution of dimethyl phosphite (22.4 mL, 244 mmol) in 400 mL of anhydrous THF at 0 "C was added NaH (9.3 g, 232 mmol, 60% dispersion in mineral oil) in small portions. (Benzy1oxy)benzyl chloride (53.7 g, 232 mmol) was then introduced via canula as a solution in 100 mL of anhydrous THF. The resulting mixture was warmed to room temperature and stirred for 18 h. The solvent was then removed in uacuo, and the resulting oil was partitioned between HzO/Et20 (300

Triacid Angiotensin 11Antagonists mL each). The layers were separated, and the aqueous layer was extracted with Et20 (2 x 200 mL). The organic was combined, dried (NazS04),and concentrated in vacuo to give 78.3 g of a thick oil. Chromatography (SiOz, 75% EtOAd25% hexane) provided 36.6 g (52%) of dimethyl (4-hydroxybenzyl)phosphonate as a solid residue; 'H N M R (CDC13) 6 7.43-7.25 (m, 5H), 7.21 (dd, J = 9.0, 3.0 Hz, 2H), 6.91 (d, J = 9.0 Hz, 2H), 5.14 (8, 2H), 3.65 (s, 3H), 3.61 (9, 3H), 3.11 (d, J = 21 Hz); FD MS 306. Anal. (C16H1904P) C, H. A solution of dimethyl [4-(benzyloxy)benzyllphosphonate (19.4 g, 63.0 mmol) in 100 mL of 1%concentrated HC1 in EtOH was treated with 840 mg of 5% PdC. The mixture was hydrogenated at 40 psi for 30 min. The reaction mixture was then filtered through a pad of Celite, and the filtrate was concentrated in vacuo t o give 13.6 g (100%) of 2e as a white solid: mp 126-129 "C; lH NMR (CDC13) 6 6.98 (dd, J = 9.0, 3.0 Hz, 2H), 6.67 (d, J = 9.0 Hz, 2H), 3.56 (8, 3H), 3.52 (9, 3H), 2.96 (d, J = 21 Hz, 2H). Anal. (CgH1304P) C, H. Diethyl (4-hydroxy-2-phenethy1)phosphonat.e (20. To a -30 "C solution of tetraethyl methylenediphosphonate (6.22 g, 21.6 mmol) in 30 mL of anhydrous THF under Nz was added n-BuLi (14.9 mL, 23.8 mmol, 1.6 M solution in hexanes) dropwise via syringe. After stirring for 30 min, 4-(benzyloxy)benzaldehyde (4.58 g, 21.6 mmol) was added as solution in 15 mL of anhydrous THF. The resulting mixture was allowed to warm to room temperature and stirred for 2 h. The reaction was quenched by pouring into HzO (200 mL). The aqueous layer was extracted with EtOAc (3 xl00 mL). The organic layer was dried (NazSOJ, and concentrated in vacuo to an oil that was chromatographed (SiOz, 1:l EtOAcihexanes) to provide 6.05 g (81%)of 4-(benzyloxy)phosphocinnamic acid diethyl ester as a colorless oil that solidified on standing: mp 43-45 "C; 1H NMR (CDC13) 6 7.52-7.32 (m, 8H), 6.97 (d, J = 8.45 Hz, 2H), 6.09 (dd, J1 = J z = 17.65 Hz), 5.10 (8, 2H), 4.15 (q, J = 7.35 Hz, 4H), 1.37 (t,J = 7.35 Hz). Anal. (C19H~304P) C, H. The above benzyl ether (6.05 g, 17.5 mmol) was dissolved in 50 mL of EtOH. To this solution was added 1.15 g of 5% P d C . The mixture was hydrogenated at 40 psi for 3 h. The reaction mixture was then filtered through a pad of Celite, and the filtrate was concentrated in vacuo to give 4.52 g (99%) of diethyl (4'-hydroxy-2-phenethyl)phosphonate (20as a colorless oil: 'H NMR (CDCl3) 6 7.04 (d, J = 8.3 Hz, 2H), 6.80 (d, J = 8.3 Hz, 2H), 4.10 (q, J = 7.1 Hz, 4H), 2.85 (m, 2H), 2.06 (m, 2H), 1.32 (t, J = 7.1 Hz, 6H); FD MS 258. Anal. (CizHi904P) C, H.

Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26 4515

Data for (2S,4S)-N-Cbz-4-[4-(2-carbethoxymethyl)phenoxy]proline methyl ester (3b): isolated in 81% yield by chromatography (SiOz, 30% EtOAdhexanes); [ah -15.8" (c 1.0, MeOH); lH NMR (DMSO-&) 6 (doubling due to amide rotamers) 7.34-7.26 (m, 5H), 7.30 (d, J = 8.4 Hz, 2H), 6.76 (d, J = 8.4 Hz, 2H), 5.12-4.96 (m, 3H), 4.53 and 4.47 (dd, J1 = 9.0,

1.5 Hz, lH), 4.02 (q, J = 7.0 Hz, 2H), 3.74 (m, lH), 3.89 and 3.54 (s,3H), 3.54 (s, 2H), 3.44-3.30 (m, lH), 2.52 (m, lH), 2.23 (m, lH), 1.14 (t, J = 7.0 Hz, 3H). Anal. (CzJI27N07) C, H, N.

Data for (2S,4S)-N-Cbz-4-[4-(2-carbethoxyethyl)phenoxylproline methyl ester (3c): isolated in 65% yield by -18.1" (c 1.0, chromatography (SiOz, 30% EtOAAexanes); [al~ MeOH); 1H NMR (DMSO-d6) 6 (doubling due to amide rotamers) 7.34-7.26 (m, 5H), 7.09 and 7.08 (d, J = 8.5 Hz, 2H), 6.72 and 6.71 (d, J = 8.5 Hz, 2H), 5.11-4.95 (m, 3H), 4.52 and 4.46 (dd, J = 9.5, 1.6 Hz, lH), 4.02 (m, 2H), 3.72 (m, lH), 3.58 and 3.53 (9, 3H), 3.46 (m, lH), 2.73 (m, 2H), 2.56-2.45 (m, 3H), 2.23 (m, lH), 1.10 (m, 3H). Anal. (Cz5HzgN07) C, H, N.

Data for (2S,4S)-N-Cbz-4-[4-(2-carbomethoxyisopropy1)phenoxylproline methyl ester (3d);isolated as an oil in 58%yield by chromatography (SiOz, 30% EtOAcihexanes); [ a ]-15.5' ~ (c 1.0, MeOH); 'H NMR (DMSO-d6) 6 (doubling due t o amide rotamers) 7.30-7.24 (m, 5H), 7.18 and 7.16 (d, J = 8.7 Hz, 2H), 6.76 (d, J = 8.7 Hz, 2H), 5.11-4.97 (m, 3H), 4.53 and 4.44 (dd, J = 8.3, 2.0 Hz, lH), 3.80-3.61 (m, lH), 3.59 and 3.54 (9, 3H), 3.53 (s, 3H), 2.55-2.46 (m, 1H), 2.252.16 (m, 1H), 1.43 (s, 6H); FD MS 455; high-resolution MS calcd for CzsHzgN07 456.2022, found 456.2013. Anal. (C25HZ9NO7)H; N; C: calcd, 65.92; found, 64.30; N: calcd, 3.08; found, 3.83.

Data for (2S,4S)-N-Cbz-4-[4-[2-(diethoxyphosphinyl)ethyllphenoxylproline methyl ester (30: isolated as an oil in 37% yield by chromatography (SiOz, 75-90% EtOAd hexanes); [ a ] -11.7' ~ (c 0.9, MeOH); 'H NMR (CDC13) 6 (doubling due to amide rotamers) 7.38-7.30 (m, 5H), 7.10 (d, J = 8.5 Hz, 2H), 6.72 (d, J = 8.5 Hz, 2H), 5.20-5.11 (m, 2H), 4.90 (m, 1H), 4.61 and 4.58 (dd, J = 6.0, 2.0 Hz, l H ) , 4.12 (q, J = 7.0 Hz, 4H), 3.82-3.76 (m, lH), 3.74 and 3.64 (s, 3H), 2.85 (m, 2H), 2.51-2.45 (m, 2H), 2.04 (m, 2H), 1.32 (t, J = 7.0 Hz, 6H). FD MS 520; high-resolution MS calcd for C~~H~ENOSP 520.2100, found 520.2135.. Anal. ( C d & f i o ~ P ) H, N; C: calcd, 60.11; found, 59.13.

Data for (2S,4S)-N-Cbz-4-[4-[ [2-(triphenylmethyl)-Wtetrazol-5-yl]methyl]phenoxy]proline methyl ester (3g): isolated in 83% yield by chromatography (SiOz, 5-25% ~ (c 1.0, MeOH); 'H NMR (DMSOEtOAdtoluene); [ a ]+9.4" General Procedure for Mitsunobu coupling of 2a-g de) 6 (doubling due to amide rotamers) 7.40-7.26 (m, 20H), to (2S,4R)-N-Cbz-4-hydroxyproline.Preparation of (2S,4S)-N-Cbz-4-[4-[(dimethoxyphosphinyl)methyllphe- 7.09 and 6.97 (d, J = 8.5 Hz, 2H), 6.97 and 6.74 (d, J = 8.5 Hz, 2H), 5.11-4.95 (m, 3H), 4.52 and 4.46 (dd, J = 6.1, 1.5 noxylproline Methyl ester (3e). To a solution of (2S,4R)Hz, lH), 4.17 (9, 2H), 3.73 (m, lH), 3.57 and 3.52 (s, 3H), 3.41 N-Cbz-4-hydroxyproline methyl ester (10.0 g, 35.8 mmol) in (m, lH), 2.53-2.46 (m, lH), 2.23 (m, 1H); FD MS 679. Anal. 400 mL of anhydrous THF under Nz at 0 "C were added (C41H37N505=0.25EtOAc(from chromatography)) C, H, N. triphenylphosphine (10.6 g, 39.4 mmol) and dimethyl-(4General Procedure for Deprotection of Proline Eshydroxybenzy1)phosphonate (7.9 g, 37.8 mmol). To this mixters. (2S,4S)-4-[4-[(Dimethoxyphosphinyl)methyllpheture was added diethyl azodicarboxylate (6.3 mL, 39.4 mmol) noxylproline Methyl Ester (4e). A solution of 3e (6.6 g, dropwise over a 30 min period. The reaction mixture was then 13.8 mmol) in 100 mL of 1%concentrated HC1 in EtOH was allowed t o warm t o room temperature and stirred for 18 h. treated with 1.0 g of 10% Pd/C. The mixture was hydrogeThe solvent was then removed in vacuo, and the residue was nated at 40 psi for 2 h and then passed through a pad of Celite chromatographed (SiOz, 50-100% EtOAc/hexane) to give 13.3 to remove the catalyst. The filtrate was concentrated in vacuo -14.2' (c 1.0, MeOH); lH g (75%) of 3e as a thick oil: [al~ to an oil and then partitioned between CHC13 and saturrated NMR (CDC13)6 (doubling due to amide rotamers) 7.35-7.28 NaHC03 (100 mL each). The layers were separated, and the (m, 5H), 7.17 (dd, J = 9.0, 3.0 Hz, 2 H), 6.72 (d, J = 9.0 Hz, organic was dried (NazS04) and concentrated in vacuo to give 2H),5.14(m,2H),4.87(m, lH),4.58and4.51(dd,J = 6 . 0 , 2 . 0 4.30 g (99%) of the crude deprotected proline ester 4e as a Hz, lH), 3.81-3.75 (m, 2H), 3.72 and 3.62 (s,3H),3.67 (s, 3H), pale yellow oil. This material was used in subsequent reac3.63 (s, 3H), 3.08 (d, J = 21 Hz, 2H), 2.49-2.40 (m, 2H). Anal. tions without further purification. [al~ -6.5" (c 1.0, MeOH); (Cz3HzsNOsP) C, H, N. Data for (2S,4S)-N-Cbz-4-(4-carbethoxyphenoxy)pro- 'H NMR (DMSO-ds) 6 7.13 (d, J = 8.5 Hz, 2H), 6.76 (d, J = 8.5 Hz, 2H), 4.80 (m, 11, 3.72 (dd, J = 9.0, 4.3 Hz, lH), 3.65 line methyl ester (3a): isolated in 83%yield by chromatog(m, 1H), 3.58 (s, 3H), 3.56 (s, 3H), 3.52 (5, 3H), 3.13 (d, J = raphy (SiOz, 30% EtOAdhexanes); [ah -42.9' (c 1.0, MeOH); 21.1 Hz, lH), 2.36 (m, lH), 1.97 (m, 1H); FD MS 343. 'H NMR (DMSO-de) 6 (doubling due t o amide rotamers) 7.86 Data for (2S,4S)-4-[4-[2-(diethoxyphospinyl)ethyllphe(d, J = 8.5 Hz, 2H), 7.33-7.26 (m, 5H), 6.92 (d, J = 8.5 Hz, noxylproline methyl ester (40:yield 75%; lH N M R (CDC13) 2H), 5.15-4.96 (m, 3H), 4.57-4.48 (m, lH), 4.23 (9,J = 7.1 6 7.09 (d, J = 8.5 Hz, 2H), 6.75 (d, J = 8.5 Hz, 2H), 4.79 (m, Hz, 2H), 3.82 -3.62 (m, lH), 3.58 and 3.53 (s, 3H), 3.51-3.46 lH), 4.10 (q, J = 6.7 Hz, 4H), 3.82 (dd, J = 9.4, 5.1 Hz, lH), (m, 2H), 2.60 (m, lH), 2.55 (m, lH), 1.26 (t, J = 7.1 Hz, 3H). 3.73 (s, 3H), 3.04 (dd, J = 12.4, 4.2 Hz, lH), 2.90-2.84 (m, Anal. (Cz3Hz~N07) C, H, N.

4516 Journal of Medicinal Chemistry, 1994, Vol. 37, No.26 2H), 2.41 (m, lH), 2.22 (m, 2H), 2.05-1.96 (m, 2H), 1.32 (t, J = 6.7 Hz, 6H); FD MS 385. Compounds 4a-d,g were taken up in anhydrous Et20 and treated with ethereal HBr until the solution was acidic (Congo red indicator). At this point, the HBr salt precipitated from solution. The solid was collected by filtration and dried in vacuo. Data for (2S,4S)-4-[4-(carbethoxy)phenoxylproline methyl ester hydrobromide (4a):yield 74%; mp 171-174 ~ (c 1.0, MeOH); lH NMR (DMSO-de) 6 9.62 (bs, "C; [ a ]+15.0° lH), 7.89 (d, J = 8.7 Hz, 2H), 6.98 (d, J = 8.7 Hz, 2H), 5.26 (m, lH), 4.69 (dd, J = 9.4, 3.5 Hz, lH), 4.24 (9, J = 7.0 Hz, 2H), 3.70 (8,3H), 3.63-3.39 (m, 2H), 2.65-2.35 (m, 2H), 1.26 (t, J = 7.0 Hz, 3H). Anal. (C15H19N05.1.OHBr) C, H, N. Data for (2S,4s)-4-[4-(carbe~o~ethyl)phenoxylproline methyl ester hydrobromide (4b): yield 92%;mp 163~ (c 1.0, MeOH); IH NMR (DMSO-de)6 9.58 165 "C; [ a ]+11.9" (bs, lH), 7.17 (d, J = 8.5 Hz, 2H), 6.81 (d, J = 8.5 Hz, 2H), 5.12 (m, 1H), 4.66 (dd, J = 9.4, 3.5 Hz, lH), 4.02 (4, J = 7.1 Hz, 2H), 3.71 (8, 3H), 3.56 (s, 2H), 3.51-3.35 (m, 2H), 2.602.33 (m, 2H), 1.14 (t, J = 7.1 Hz, 3H). Anal. (C16H21N05-1.0HBr) C, H, N. Data for (2S,4S)4-[4-(2-~arbethoxyethyl)phenoxy3proline methyl ester hydrobromide (4c): yield 81%;mp 117120 "C; [ a ]+10.7' ~ (c 1.0, MeOH); lH NMR (DMSO-de)6 9.40 (bs, lH), 7.12 (d, J = 8.5 Hz, 2H), 6.76 (d, J = 8.5 Hz, 2H), 5.10(m,IH),4.66(dd, J = 9 . 4 , 3 . 4 H z , l H ) , 3 . 9 9 ( q , J = 7 . 1 Hz, 2H), 3.70 (5, 3H), 3.62-3.39 (m, 2H), 2.74 (t,J = 7.4 Hz, 2H), 2.53 (t, J = 7.4 Hz, 2H), 2.50-2.32 (m, 2H), 1.10 (t, J = 7.1 Hz, 3H). Anal. (C1.iHz3NOgl.OHBr) C, H, N. Data for (2S,4S)-4-[4-(2-~arbomethoxyisopropyl)phenoxylproline methyl ester hydrobromide (4d):yield 57%; [ a ] +10.5" ~ (c 1.0, MeOH); mp 99-103 "C; 'H NMR (DMSOde) 6 9.40 (bs, lH), 7.21 (d, J = 8.6 Hz, 2H), 6.82 (d, J = 8.6 Hz, 2H), 5.11 (m, lH), 4.65 (dd, J = 9.4, 3.4 Hz, W , 3.71 (s, 3H), 3.54 ( 8 , 3H), 3.50-3.22 (m, 2H), 2.61-2.32 (m, 2H), 1.44 (9, 6H). Anal. ( C ~ , H Z ~ N O ~ ~ . O C,HH, B ~N.) Datafor (2S,4S)-4-[4-(W-tetrazol-S-ylmethyl)phenoxyIproline methyl ester hydrobromide (4g): yield 77%; [ab +8.6" (c 1.1,MeOH); mp 150-155 "C dec; 'H NMR (DMSOde) S 9.89 (bs, lH), 9.23 (bs, lH), 7.18 (d, J =8.5 Hz, 2H), 6.83 (d, J = 8.5 Hz, 2H), 5.12 (m, 2H), 4.66 (m, lH), 4.18 (s, 2H), 3.70 (s, 3H), 3.54-3.39 (m, 2H), 2.59-2.31 (m, 2H); FD MS 304. Anal. (C14H17N503.1.5HBr) C, H, N. (f)-4-Nitroimidazole-2-octanoic Acid (5). To a suspension of NaH (17.5 g, 0.44 mol, 60% dispersion in mineral oil) in anhydrous DMF (300 mL) under N2 was added 4-nitroimidazole (49.5 g, 0.44 mol) in small portions such that the internal temperature did not rise above 30 "C. After gas evolution ceased, ethyl 2-bromooctanoate (107 g, 0.426 mol) was introduced dropwise via a n addition funnel. After stirring for 2 h at room temperature, the reaction was poured into icewater (1 L) and extracted with EtOAc (3 x 500 mL). The organic was dried (Na2S04) and concentrated in vacuo t o an oil. This material was passed through a pad of Si02 using 1:l hexanes/EtOAc as eluant. Concentration provided 125.7 g (loo%, contains residual EtOAc) of (f)-ethyl4-nitroimidazole2-octanoate as a light yellow liquid. This material was used in the next reaction without further purification. (&)-Ethyl4-nitroimidazole-2-octanoate(125.7 g crude, 0.425 mol) was dissolved in 120 mL of EtOH. To this solution were added 1.06 L of 2 N NaOH and 100 mL of THF. The resulting mixture was stirred at room temperature for 2 h. The reaction mixture was extracted with Et20 (2 x 300 mL). The aqueous was then acidified to pH = 3.6 with 5 N HC1. The aqueous layer was extracted with EtOAc (3 x 300 mL). The organic layer was dried (NazS04)and concentrated in vacuo to provide 97.2 g (90%)of 5 as a thick oil that solidified on standing: 'H NMR (CDCl3) 6 8.97 (bs, lH), 7.91 (s, lH), 7.72 (8,lH), 4.80 (dd, J = 9.0, 6.0 Hz, lH), 2.29 (m, lH), 2.03 (m, lH), 1.291.19 (m, 8H), 0.85 (t,J = 6.0 Hz, 3 H). Anal. (CllH17N304)C, H, N. Resolution of 5. A mixture of (f)-S (28.75 g, 112 mmol), (-)-cinchonidine (16.5 g, 56 mmol), and triethylamine (5.69 g, 56 mmol) in 330 mL of 1:2 EtOWH20 was heated under reflux until a solution was obtained. The solution was allowed

Palkowitz et al. to cool and stirred at room temperature for 24 h. The product was collected by filtration, washed with 1:2 EtOWHzO (2 x 150 mL), and dried, affording 25.12 g of (R)-2-(4-nitro-l.Himidazol-1-y1)-octanoicacid-cinchonidinesalt as colorless crystals (91.5%e.e.). The product was recrystallized from 330 mL of 1:2 EtOWHzO to give 22.3 g (72%, '99% e.e.). A sample of the free acid was generated by partitioning 2.00 g (3.63 mmol) of the salt between 30 mL of EtOAc and 30 mL of 1N HC1. The organic phase was washed with 10 mL of brine, dried (MgS04),and concentrated in vacuo to give 0.93 g (100%) of (R)-S as a n off-white solid. Data for (R)-2-(4-nitro-l.H-imidazol-l-yl)octanoic acid cinchonidinesalt: mp 205 "C dec; [ a ]-111.1" ~ (c 1.0, EtOH); 'H NMR (CDCl3) 6 8.81 (d, J = 4.4 Hz, lH), 8.06 (d, J = 8.1 Hz, lH), 7.98 ( 8 , lH), 7.90 (d, J = 8.3 Hz, lH), 7.67 (m, 2H), 7.53 (9, lH), 7.43 (m, lH), 6.40 (bs, lH), 6.23 (8,lH), 5.54 (m, lH), 5.00 (m, 2H), 4.55 (dd, J = 10.0, 5.0 Hz, lH), 4.28 (m, lH), 3.34 (m, 2H), 3.18 (m, lH), 2.97 (m, lH), 2.63 (m, lH), 2.22 (m, lH), 2.00 (m, 5H), 1.76 (m, lH), 1.24 (m, 9H), 0.83 (t, J = 6.6 Hz, 3H). Anal. ( C ~ O H ~ ~ NC,SH, O ~N.) Data for (R)-5: '99% ee; mp 116-118 'C; [ab -32.5' (c 1.0, EtOH); IH NMR (same as for racemate). Anal. (CllH17N304) C, H, N. Method for Determination of Enantiomeric Excess (ee). The free acid was esterified with diazomethane in Et20 and analyzed by chiral HPLC. Analysis conditions: Chiralcel OD column, 8 5 1 5 hexane/isopropyl alcohol, flow rate 1 m U min, 1 = 282 nm. tR: (S)-S,5.9 min; (R)-S,9.0 min. General Method for the Synthesis of 6a-e and 7a-e. Preparation of cb4-[4I(Dimethoxyphosphinyl)methyllphenoxy]-(R)-l-[l-oxo-2-(4-nitrol.H--imidazol-l-yl)octyllL-proline Methyl Ester (6e) and cb4-[4-[(Dimethoxyphosphinyl)methyl]phenogyl-(S)-l-[l-oxy-2-(4-nitro-l.Himidazol-1-yl)octyl]-~-proline Methyl Ester (7e). (&)-E (3.7 g, 14.5 mmol) was dissolved in 25 mL of anhydrous CHZClz. To this solution was added oxalyl chloride (1.7 mL, 18.9 mmol) followed by 3 drops of DMF. When gas evolution ceased, the solvent was removed in vacuo to give the acid chloride as an amber oil that was evaporated from an additional 20 mL of CHzC12. The acid chloride was used immediately in the next reaction. To a solution of 4e in 20 mL of anhydrous CHzClz at 10 "C was added N,N-diisopropylethylamine (2.7 mL, 15.1 mmol). The acid chloride was then introduced dropwise from an addition funnel as a solution in 10 mL of CHzC12. The resulting mixture was warmed to room temperature and stirred for 18 h. The reaction mixture was next distributed between EtOAc/HzO (200 mL ea.). The layers were separated, and the aqueous layer was extracted with EtOAc (3 x 100 mL). The organic layer was combined and washed with brine followed by HzO. The organic was then dried (Na2SOd and concentrated in vacuo to give an oil. The diastereomeric octanoamides were separated by chromatography (SiOz, 1% MeOWEtOAc) t o give 1.57 g of the (R,S,S) isomer 6e (first isomer to elute) and 1.25 g of the (S,S,S) isomer 7e, along with 1.12 g of a mixed fraction that was rechromatographed t o provide an additional 480 mg of 6e and 565 mg of 7e. Yield of 6e is 50%. Yield of 7e is 44%. -63.5' (c 1.0, MeOH); Rf 0.27 (955, Data for 6e: [ a l ~ EtOAdMeOH); 'H NMR (DMSO&) 6 (doubling due t o amide rotamers) 8.45 and 8.44 (8, lH), 8.00 and 7.96 (8, 3H), 7.15 and 7.14 (d, J = 8.5 Hz, 2H), 6.77 and 6.75 (d, J = 8.5 Hz, 2H), 5.33 and 5.18 (dd, J1 = JZ = 7.5 Hz, lH), 5.13 and 4.59 (dd, J = 9.3, 1.7 Hz, lH), 5.11 and 5.00 (m, lH), 4.01-3.41 (m, 2H), 3.64 and 3.50 (s, 3H), 3.57 (9, 3H), 3.53 (9, 3H), 3.14 (d, J = 21.1 Hz, 2H), 2.53-2.14 (m, 2H), 2.10-1.95 (m, 2H), 1.16-0.80 (m, l l H ) , 0.19 (m, 3H); FD MS 580. Anal. (C2eH37N409P) C, H, N. Data for 7e: mp 71-75 "C; [ a ]+39.8" ~ (c 1.0, MeOH); Rf 0.20 (9.55, EtOAdMeOH); 'H NMR (DMSO-ds) 6 (doubling due to amide rotamers) 8.43 and 8.30 (s, lH), 7.94 and 7.78 (8,lH), 7.18 and 7.13 (d, J = 8.3 Hz, 2H), 6.82 and 6.71 (d, J = 8.3 Hz, 2H), 5.33 and 5.23 (dd, J1 = JZ = 7.4 Hz, lH), 5.14 and 5.02 (m, lH), 4.92 and 4.59 (dd, J = 9.1, 2.1 Hz, lH), 4.203.89 (m, lH), 3.60 and 3.58 (s, 3H), 3.56 and 3.55 (s, 3H), 3.53 and 3.51 (s, 3H), 3.32 (m, lH), 3.16 and 3.08 (d, J = 21.1 Hz,

Triacid Angiotensin 11Antagonists

Journal of Medicinal Chemistry, 1994, Vol. 37,No. 26 4517

2H), 1.21-0.95 (m, 11H), 0.79 (m, 3H); FD MS 580. Anal. (Cz6H37N409P) c , H, N.

Data for cis-4-(4-carbethoxyphenoxy)-l-[ l-ox0-2(R)-(4nitro-1H~imidazol-l-yl)octyl]-~-proline methyl ester (6a): isolated as a solid in 78%yield by chromatography (Si02, 30-60% EtOAdhexanes); [ a l -81.0' ~ (c 1.0, MeOH); Rf 0.30 (7525, EtOAdhexanes); lH NMR (DMSO-ds) 6 (doubling due to amide rotamers) 8.45 and 8.43 (8,lH), 7.99 and 7.94 (9, lH), 7.87 (d, J = 8.6 Hz, 2H), 6.93 (d, J = 8.6 Hz, 2H), 5.32 and 5.22 (dd, J1 = JZ = 7.3 Hz, lH), 5.20 and 4.62 (dd, J = 9.0, 1.5 Hz, 2H), 5.18 and 5.15 (m, lH), 4.24 (q, J = 7.1 Hz, 2H), 4.03 and 3.70 (dd, J = 11.0, 4.0 Hz, lH), 3.86 and 3.44 (J= 11.0 Hz, 1.5 Hz, lH), 3.63 and 3.50 (9, 3H), 2.56-2.17 (m, 2H), 2.02-5.95 (m, 2H), 1.24-0.95 (m, l l H ) , 0.79 (m, 3H). Anal. (C26H3a408) C, H, N.

Data forcis-4-(4-carbethoxyphenoxy)-l-[ i-oxo-2(S)-(4nitro-~-imidazol-l-yl)octyl]-~-p~line methyl ester (7a): isolated as a solid in 81% yield by chromatography (SiOz, 3060% EtOAdhexanes); mp 100-106 'C; [ah f47.4" (c 1.0, MeOH); Rf 0.14 (7525, EtOAdhexanes); 'H NMR (DMSO-&) 6 (doubling due to amide rotamers) 8.44 and 8.31 (9, lH), 7.94 and 7.79 (9, lH), 7.90 and 7.84 (d, J = 8.5Hz, 2H), 6.98 and 6.88 (d, J = 8.5 Hz, 2H), 5.27 and 5.24 (dd, J1 = Jz = 7.8 Hz, IH), 5.16 and 4.95 (m, lH), 5.26 and 4.63 (dd, J1 = 8.4 Hz, J 2 = 1.5 Hz, lH), 4.28-4.16 (m, 2H), 3.73-3.48 (m, 2H), 3.59 (8, 3H), 2.58-2.20 (m, 2H), 1.95 (m, 2H), 1.29-0.95 (m, 11Hh 0.81-0.72 (m, 3H); FD MS 531. Anal. (Cz6H38408)C, H, N.

2.70 (m, 2H), 2.56-2.14 (m, 4H), 2.05-1.95 (m, 2H), 1.170.94 (m, 11H), 0.78 (m, 3H). Anal. (CzsH38N408) C, H, N.

Data for ci~-4-[4-(carbomethoxyisopropyl)phenoxyl1-[l-oxo~2(R)-(4-nitro-1W-imidazol-l-yl)octyll-~-proline methyl ester (6d): isolated as an oil in 55% yield by chromatography (SiOz, 30-60% EtOAdhexanes); [aln -57.2" (c 1.0, MeOH); Rf 0.36 (75:25 EtOAdhexanes); lH NMR (DMSO-d6) 6 (doubling due to amide rotamers) 8.45 and 8.43 (s, lH), 7.99 and 7.95 (s, lH), 7.18 (d, J = 8.4 Hz, 2H), 6.77 and 6.55 (d, J = 8.4 Hz, 2H), 5.32 and 5.29 (dd, J1 = JZ = 7.5 Hz, lH), 5.00 and 4.99 (m, lH), 5.11 and 4.59 (dd, J = 8.2, 1.5 Hz, lH), 3.83-3.40 (m, 2H), 3.63 and 3.51 (s,3H), 3.54 (s, 3H), 2.54 (m, lH), 2.13-1.98 (m, 3H), 1.44 (9, 6H), 1.18-0.97 (m, lH), 0.78 (m, 3H); FD MS 588. Anal. (Cz~H38N408)C, H, N. Data for cis-4-[4-(carbomethoxyisopropyl)phenoxyl1-[l-oxo-2(S)-(4-nitro-W-imidazol-l-yl)octyl]-~-proline methyl ester (7d): isolated as a n oil in 53% yield by chromatography (SiOz, 30-60% EtOAdhexanes); [aln +31.1" (c 1.0, MeOH); Rf 0.26 (7,525, EtOAJhexanes); 'H NMR (DMSO-d6) 6 (doubling due t o amide rotamers) 8.43 and 8.30 (s, lH), 7.94 and 7.78 (8, lH), 7.20 and 7.15 (d, J = 8.7 Hz, 2H), 6.83 and 6.71 (d, J = 8.7 Hz, 2H), 5.32 and 5.29 (dd, J1= Jz = 7.6 Hz, lH), 5.14 and 5.02 (m, lH), 4.93 and 4.60 (dd, J = 9.5, 3.0 Hz, lH), 4.19 and 3.64 (m, lH), 3.48-3.45 (m, lH),

3.60and3.59(~,3H),3.54and3.52(~,3H),2.55(m,lH),2.19-

1.92 (m, 3H), 1.44 and 1.42 (s,6H), 1.28-1.07 (m, 11H), 0.78 (transfer (m, 3H); FD MS 588. Anal. (C~sH38N408.0.40CHzCl~ Data for cis-4-[4-(carbethoxymethyl)phenoxyl-l-[l- solvent)) C, H, N.

oxo-2(R)-(4-nitro-~-imidazol-l-yl)odyll-~-proline methyl ester (6b): isolated as an oil in 79% yield by chromatog-

raphy (SiOZ, 30-60% EtOAdhexanes); [ a l ~ -67.9" (c 1.0, MeOH); Rf 0.49 (7525 EtOAdhexanes); 'H NMR (DMSO&) 6 (doubling due to amide rotamers) 8.45 and 8.44 (9, lH), 7.99 and 7.95 (s, lH), 7.14 (d, J = 8.4 Hz, 2H), 6.77 and 6.75 (d, J = 8.4 Hz, 2H), 5.33 and 5.22 (dd, J1 = J 2 = 7.4 Hz, lH), 5.19 and 4.59 (dd, J =8.5, 1.4 Hz, lH), 5.14 and 5.01 (m, lH), 4.06 and 3.59 (m, lH), 4.03 (9, J = 7.1 Hz, 2H), 3.83 and 3.42 (dd, J = 11.2 Hz, 1.5 Hz, lH), 3.65 and 3.51 (8,3H), 3.54 (9, 2H), 2.53-1.96 (m, 4H), 0.80 (m, 3H); FD MS 544. Anal. (Cz~H36N408) C, H, N.

c~~[4(W-Tetrazol-~y~e~yl)phenoxyl-l-[1-0~~2~)(4-nitro-W-imidazol-l-yl)octyll-~-proline Methyl Ester (6g). To a solution of 4g (0.58 g, 1.38 mmol) in 13.0 mL of DMF at room temperature was added N,N-diisopropyleth-

ylamine (0.72 mL, 4.14 mmol). After stirring for 30 min, (R)-5 (0.39 g, 1.52 "01) was added along with hydroxybenzotriazole (HOBT) (0.20 g, 1.52 mmol). After stirring an additional 10 min, dicyclohexylcarbodiimide (DCC) (0.37 g, 1.79 mmol) was added in small portions. The resulting mixture was stirred for 16 h after which time the reaction mixture was diluted with EtOAc (75 mL) and the precipitated dicyclohexylurea (DCU) was removed by filtration. The filtrate was washed Data for cis-4-[4-(carbethoxymethyl)phenoxyl-l-[1- with HzO (4 x 50 mL). The organic was dried (NazS04) and concentrated in vacuo to an oil that was chromatographed oxo-2(S)-(4-nitroW-imidazoll-yl)octyll-~-proline meth(SiOz, 98:2 CHClJMeOH) to provide 570 mg (76%) of 6g as a yl ester (7b):isolated as a solid in 74% yield by chromatogcolorless oil: [ a l ~ -58.9" (c 1.0, MeOH); 'H NMR (DMSO-&) raphy (SiOz, 30-60% EtOAdhexanes); mp 69-73 "C; [aln 6 (doubling due to amide rotamers) 8.45 and 8.42 (s, lH), 7.99 +31.1" (c 1.0, MeOH); Rf0.24 (7525 EtOAdhexanes); 'H N M R and 7.94 (s, lH), 7.15 (d, J = 8.4 Hz, 2H), 6.78 and 6.77 (d, J (DMSO-ds)6 (doubling due t o amide rotamers) 8.44 and 8.30 = 8.4 Hz, 2H), 5.32 and 5.18 (dd, J1 = JZ= 7.3 Hz, lH), 5.17 (5, lH), 7.95 and 7.79 (9, lH), 7.16 and 7.12 (d, J = 8.4 Hz, and 4.58 (dd, J =8.3, 1.5 Hz, lH), 5.12 and 5.01 (m, lH), 4.17 2H), 6.82 and 6.71 (d, J = 8.4 Hz, 2H), 5.32 and 5.23 (dd, Ji = and 3.64 (s,3H), 4.00 and 3.68 (dd, J = 11.2,4.5 Hz, lH), 3.81 Jz = 7.5 Hz, lH), 5.15 and 5.03 (m, lH), 4.92 and 4.60 (dd, J and 3.43 (11.2, 1.5 Hz, lH), 2.85 and 2.69 (9, 2H), 2.52-2.16 = 8.9,2.3 Hz, lH), 4.20-3.98 (m, 3H), 3.68-3.62 (m, lH), 3.59 (m, 3H), 2.07-1.87 (m, 2H), 1.77-0.99 (m, 8H), 0.79 (m, 3H); and 3.58 (9, 3H), 3.55 and 3.52 (s, 2H), 2.55-2.15 (m, 2H), FD MS 540. Anal. ( C ~ S H ~ Z N0.30CHC13 ~OC (from chromatog1.97-1.93 (m, 2H), 1.20-0.99 (m, l l H ) , 0.78 (m, 3H); FD MS raphy)) C, H, N. 544. Anal. (C2&&08) c , H, N.

Data for cis-4-[4-(2-carbethoxyethyl)phenoxyI-l-[1o x o - ~ ( R ) - ( 4 - n i t r o - W - ~ i d a z o l - l - y l ) o cmethtyle yl ester (64:isolated as a n oil in 74% yield by chromatography (SiOz, 30-60% EtOAdhexanes); [a10 -62.2" (c 1.0, MeOH); Rf 0.41 (7525, EtOAdhexanes); 'H NMR (DMSO-ds) 6 (doubling due t o amide rotamers) 8.45 and 8.42 (8,lH), 7.99 and 7.94 (s, lH), 7.09 (d, J = 8.2 Hz, 2H), 6.72 and 6.71 (d, J = 8.2 Hz, 2H), 5.33 and 5.17 (dd, J1 = JZ= 7.5 Hz, lH), 5.09 and4.99(m, lH),5.15and4.58(dd,J=8.7,1.5Hz,lH),4.00 (9, J = 7.0 Hz, 2H), 3.81-3.40 (m, 2H), 3.65 and 3.50 (s, 3H), 2.74 (t, J = 7.4 Hz, 2H), 2.53 (t, J = 7.4 Hz, 2H), 2.41-2.03 (m, 2H), 1.99-1.94(m, 2H), 1.27-0.94 (m, 11H), 0.78(m, 3H). Anal. ( C ~ S H ~ ~ NC, ~O H,SN. ) Data for cis-4-[4-(2-carbethoxyethyl)phenoxyI-l-[1oxo-2(S)-(4-nitro-W--imidazoll-yl)octyll-~-proline methyl ester (7c): isolated as an oil in 63% yield by chromatography (Si02, 30-60% EtOAdhexanes); [ a l ~+32.3" (c 1.0, MeOH); Rf 0.28 (7525 EtOAdhexanes); 'H NMR (DMSO-&) 6 (doubling due t o amide rotamers) 8.43 and 8.30 (s, lH), 7.94 and 7.78 (9, IH), 7.12 and 7.07 (d, J = 8.4 Hz, 2H), 6.78 and 6.66 (d, J = 8.4 Hz, 2H), 5.31 and 5.22 (dd, J1 = JZ = 7.5 Hz, lH), 5.12 and 5.00 (m, lH), 4.91 and 4.59 (dd, J = 8.0, 1.5 Hz, lH), 4.17-3.95 (m, 3H), 3.66-3.43 (m, lH), 3.59 (s, 3H), 2.77-

Data for cis-4-Phenoxy-l-[l-oxo-2(R)-(4-nitro-W-imi-

dazol-1-yl)octyl]- proli line Methyl Ester (6h)and cis-4-

Phenoxy-1-[ l-oxo-2(S)-(4-nitro-W-imidazol-l-yl)octyll-~proline Methyl Ester (7h). Compounds 6h and 7h were prepared exactly as described for Ga-d,e and 7a-d,e. Data for 6h: isolated by chromatography (SiOz, hexanes/EtOAc) as a semisolid in 78% yield; [ a ]-64.1' ~ (c 1.0, MeOH); 'H NMR (DMs0-d~)6 (doubling due to amide rotamers) 8.45 and 8.43 (5, lH), 7.99 and 7.95 (8, lH), 7.25 and 6.92 (m, 3H), 6.816.79 (m, 2H), 5.32 and 5.19 (dd, J1 = JZ = 7.2 Hz, lH), 5.18 and 4.59 (dd, J = 11.3, 1.7 Hz, lH), 5.14 and 5.03 (m, lH), 4.02 and 3.63 (dd, J = 11.3, 3.5 Hz, lH), 3.82 and 3.42 (dd, J = 11.3, 1.0 Hz, lH), 3.65 and 3.51 (s, 3H), 2.46-1.95 (m, 4H), 1.18-0.92 (m, 8H), 0.79 (m, 3H). Anal. (C~3H30N406) C, H, N. Data for 7 h isolated by chromatography (SiOz, hexanes/ EtOAc) as a white solid in 64% yield; recrystallized from hexanes/EtOAc; mp 98-100 "C; [ a l $34.4" ~ (c 1.0, MeOH); 'H NMR (DMSO&) 6 (doubling due to amide rotamers) 8.44 and 8.30 (8, lH), 7.94 and 7.79 (s, lH), 7.21 and 6.95 (m, 3H), 6.88 and 6.76 (d, J = 8.1 Hz, 2H), 5.32 and 5.20 (dd, JI = JZ = 7.2 Hz, lH), 5.19 and 5.05 (m, lH), 4.93 and 4.60 (dd, J =9.1, 1.2 Hz, lH), 4.17 and 3.48 (dd, J = 11.3, 4.6 Hz, lH), 3.65 and

Palkowitz et al.

4518 Journal of Medicinal Chemistry, 1994, Vol. 37,No. 26

solid. mp 150-155 "C; 'I3 NMR (DMSO-de) 6 (doubling due to amide rotamers) 11.81and 11.68 (s, lH), 7.99 (bs, lH), 7.88 (m, lH), 7.68-7.40 (m, lH), 7.61 and 7.57 (s, lH), 7.46 (m, 2H), 7.14 (d, J = 8.6 Hz, 2H), 6.76 and 6.72 (d, J = 8.6 Hz, General Method for the Preparation of 8a-e,g. Preparation of cie-4-[4-[(Dimetho~hosphinyl)methyllphe- 2H), 5.21-4.98 (m, 2H), 5.03 and 4.59 (dd, J = 9.4, 1.9 Hz, noxyl-1-[1-oxo-2(R)-[4-[(2-sulfobenzoyl)aminol-1H-imi- lH), 4.05-3.49 (m, 2H), 3.92 (9, J = 7.0 Hz, 4H), 3.94 and 3.67 (s, 3H), 2.67 (m, 2H), 2.51-2.10 (m, 4H), 2.04-1.88 (m, dazol-l-yl]octyl]-~proline Methyl Ester (8e). To a solution of 8e (4.00 g, 6.90 mmol) in 50 mL of absolute EtOH was added 2H), 1.20-1.01 (m, 14H), 0.82 (m, 3H). Anal. (C36H49N40111.0 g of 5% Pd/C. The mixture was hydrogenated at 40 psi PS.0.8MeOH) C, H, N. for 30 min. The catalyst was then removed by passing the General Method for the Preparationof la-d,g. Prepamixture through a pad of Celite. The filtrate was concentrated ration of cis-4-[4-(Carboxymethyl)phenoxyl-l-[l-oxoto an amber oil that was evaporated twice from anhydrous 2(R)-[4-[(2-sulfobenzoyl)aminol-UZ-~idazol-l-ylloctyllTHF. L-proline(lb). To a solution of 8b (1.10 g, 1.60 mmol) in 100 In a separate flask, sulfobenzoic anhydride (1.40 g, 7.60 mL of THF at room temperature was added 1 N NaOH (5.0 mmol) was dissolved in 5 mL of anhydrous THF under Nz. To mL). The reaction mixture was stirred for 3 h, after which this solution was added the above aminoimidazoleas a solution time the THF was removed in uucuo. The aqueous layer was in 5 mL of anhydrous THF. After stirring for 30 min, the diluted with HzO (10 mL) and acidified to pH = 1.5 using 5 N solution was triturated with EhOhexanes to yield 4.70 g (93%) HC1. The aqueous was extracted extracted with 9O:lO EtOAd of the sulfonic acid 88 as a light yellow solid that was collected EtOH (3 x 15 mL). The organic was dried (NazS04) and by filtration. This product was used in the next reaction concentrated in uacuo to a solid that was triturated from CH3without further purification or characterization: mp 110 "C CN/EbO. Vacuum filtration provided 610 mg (58%)of lb as dec;FAB MS 735.2. a white solid: mp 160-175 "C; [al~ -14.1' (c 1.0, MeOH); 'H Preparation of (2(R)-[4-[(2-Sulfobenzoyl)amino]-UZ- NMR (DMSO-ds) 6 (doubling due to amide rotamers) 12.00 imidazol-1-ylloctanoic Acid (9). (R1-S (16.0 g, 63.0 mmol) and 11.90 (s, lH), 8.50 (bs, lH), 7.87-7.41 (m, 5H), 7.14 and was dissolved in 1.0 L of anhydrous MeOH along with 300 7.13 (d, J = 8.2 Hz, 2H), 6.81 and 6.76 (d, J = 8.2 Hz, 2H), The reaction was heated to reflux for 16 h. mg of pTsOH. 5.35 and 5.13 (dd, J1 = JZ = 7.2 Hz, lH), 5.11 and 4.96 (m, Upon cooling, the solvent was removed in uucuo t o give a n oil lH), 5.02 and 4.45 (d, J = 9.4,2.0 Hz, lH), 3.45 (s, 2H), 4.02 that was partitioned between EtOAdsaturated NaHC03 soluand 3.73 (dd, J = 11.4, 3.5 Hz, lH), 3.89 and 3.38 (dd, J = tion (300 mL each). The layers were separated, and the 11.4, 1.0 Hz, lH), 2.55-1.89 (m, 4H), 1.18-0.96 (m, 8H), 0.81 organic layer was dried (Na~S04)and concentrated in uacuo (m, 3H). Anal. ( C ~ I H ~ ~ N ~C, O~ H,ON. S) to provide 13.2 g (78%) of (R)-methyl [4-[(2-sulfobenzoyl)-

3.48 (dd, J = 11.3, 4.6 Hz, lH), 3.59 (s, 3H), 2.56-1.95 (m, 4H), 1.19-0.90 (m, 8H), 0.78 (m, 3H). Anal. (C2&&06) C, H, N.

aminol-1H-imidazol-l-yllodanoateas an amber oil: [al~ -16.8" (c 1.0, MeOH); 'H NMR (CDCl3) 6 7.91 (s, lH), 7.53 (s, lH), 4.74 (dd, J = 7.1, 4.5 Hz, lH), 3.81 (s, 3H), 2.22 (m, lH), 2.03

Data for cie-4-(4-carboxyphenoxy)~l-[l-oxo-2(R)-[4-[(2sulfobenzoyl)aminol-UZ-imidazol-l-ylloctyll-~-proline -6.2" (c 1.0, (la): isolated in 61% yield; mp 185-195 "C; [al~ MeOH); 'H NMR (DMSO-ds) 6 (doubling due to amide rota-

(m, lH), 1.20-1.18 (m, 8H), 0.84 (t, J = 7.1 Hz, 3H); FD MS mers) 12.05 and 11.93 (8, lH), 8.59 (bs, lH), 7.87-7.46 (m, 269. Anal. (C12H19N304) C, H, N. 7H), 6.95 and 6.88 (d, J = 8.6 Hz, 2H), 5.41 and 5.16 (dd, J1 = A sample of the ester was hydrolyzed to the acid with 2 Jz = 7.2 Hz, lH), 5.24 and 5.09 (m, lH), 5.10 and 4.47 (dd, J equiv of NaOH in MeOWHz0. This material was determined = 9.4 Hz, 1.0 Hz, lH), 4.05 and 3.74 (dd, J = 13.2, 4.6 Hz, t o be 98% ee using the analytical method described earlier. lH), 3.97 and 3.50 (dd, J = 13.2, 1.0 Hz, lH), 2.58-1.93 (m, To a solution of the ester (13.0 g, 45.7 mmol) in EtOH (150 4H), 1.31-0.97 (m, 8H), 0.80 (m, 3H). Anal. (C30H34N4010S) mL) was added 10% Pd/C (2.0 g). The mixture was hydrogeC, H, N. nated at 40 psi for 2 h. The catalyst was removed by passing the mixture through a pad of Celite. The filtrate was Data for cis-4-[4-(2-carboxyethyl)phenoxyl-l-[l-oxoconcentrated t o a yellow oil and then dissolved in anhydrous 2(R)-[4-[(2-sulfobenzoyl)aminolUZ-imidazol-l-ylloctyllTHF (100 mL). To this solution was added KOAc (4.44 g, 45 L-proline(IC): isolated in 48% yield; mp 140-148 "C; [al~ mmol), KzC03 (3.12 g, 22.5 mmol) followed by sulfobenzoic -14.4' (c 1.0, MeOH); 'H NMR (DMSO-de) 6 (doubling due to anhydride (8.83 g, 47.7 mmol). After stirring for 4 h, a white amide rotamers) 12.04 and 11.92 (s, lH), 8.57 (bs, lH), 7.86precipitate formed. The mixture was diluted with THF (100 7.52 (m, 3H), 7.50 (m, 2H), 7.16 (d, J = 8.5 Hz, 2H), 6.78 and mL) and the solid collected by filtration to provide 22.5 g of 6.73 (d, J = 8.5 Hz, 2H), 5.38 and 5.16 (dd, JI = JZ = 7.4 Hz, ~ 5.06 and 5.03 (m, lH), 5.02 and 4.46 (d, J = 11.1, 1.5 Hz, crude ~ ( R ) - [ 4 [ ( 2 - 8 u l f o b e n z o y l ) a m i n o l - ~ - ~ d a z o l - l - y l l ~ o alH), potassium salt. This material was dissolved in a mixture of lH), 4.04 and 3.65 (m, lH), 3.99 and 3.46 (dd, J = 11.1Hz, HzO (200 mL) and EtOH (100 mL). To this solution was added 1.5 Hz, lH), 2.70 (m, 2H), 2.46-1.87 (m, 6H), 1.35-0.91 (m, 1N NaOH (53 mL). After stirring at room temperature for 3 8H), 0.81 (m, 3H). Anal. ( C ~ Z H ~ S N ~ OC,I OH,S )N. h, the solution was concentrated in uucuo to remove the EtOH. Data for cis-4-[4-(carbomethoxyisopropyl)phenoxylThe aqueous solution was then acidified to pH = 1.5 with 5 N 1-[1-oxo-2 (R)[4-[(2-sulfobenzoyl)aminol -W-imidazol-1 HC1. This solution was then extracted with 9:l EtOAc/EtOH yl]octyl]- proli line (Id): isolated by filtration of aqueous (3 x 200 mL). The organic was dried (NazS04) and concensolution in 88% yield; mp 182-187 "C; [al~-29.4 " (c 0.5, trated in uacuo to give 8.65 g (46% for two steps) of 9 as a MeOH); IH NMR (DMSO-ds) 6 (doubling due to amide rotawhite solid: mp 250-260 "C dec; [al~ -8.3" (c 1.0, MeOH); 'H mers) 12.10 and 11.98 (s, lH), 8.70 (bs, lH), 7.86-7.48 (m, NMR (DMSO-de) 6 12.07 (s, lH), 8.63 (s,lH), 7.86 (d, J =6.4 5H), 7.24 and 7.21 (d, J = 8.5 Hz, ZH), 6.83 and 6.78 (d, J = Hz, lH), 7.67 (m, 2H), 7.54-7.49 (m, 2H), 5.16 (dd, J1 = JZ = 8.5 Hz, 2H), 5.40 and 5.18 (dd, J1 = JZ = 7.4 Hz, lH), 5.10 and 7.4 Hz, lH), 2.10 (m, 2H), 1.20-1.09 (m, 8H), 0.81 (m, 3H). 4.96(m,lH),5.08and4.46(dd,J=9.0,1.2Hz,lH),4.01and H,S N. Anal. ( C I ~ H Z ~ N ~C, O~ ) 3.69 (dd, J = 13.0 Hz, 3.5 Hz, lH), 3.93 and 3.46 (d, J = 13.0 Preparationof cis-4-[4-[2-(Diethosyphosphinyl)ethyl]- Hz, 1.2 Hz, lH), 2.54-1.96 (m, 4H), 1.41 (8, 6H), 1.36-1.08 phenoxyl-l-[l-oxo-2~R~-[4-[~2-sulfobenzoyl~aminol-~(m, 8H), 0.81 (m, 3H). Anal. (C33H~N4010S-O.6HCl) C, H, N. imidazol-1-yl]octyll-~-proline Methyl Ester (80. To a Data for cis-4-[4-(2H-tetrazol-5-ylmethyl)phenoxyl-lsolution of 4f (1.88 g, 4.90 mmol) and 9 (2.00 g, 4.90 mmol) in [l-oxo-2(R)-~4-[(2-sulfobenzoyl~aminol-~-~idazoll-yll15.0 mL of dry DMF under Nz was added HOBT (0.73 g, 5.4 octyll- proli line (lg): isolated by filtration of aqueous solummol). After the mixture was stirred for 10 min, DCC (1.13 tion in 52% yield; a sample suitable for characterization was g, 5.4 mmol) was added in small portions. The resulting purified by reverse-phase (CIS)HPLC (1%HOAd30% CH3CN/ mixture was stirred at room temperature for 18 h. The 60% HzO); mp 165-172 "C; [ a ]-19.6" ~ (c 1.0, MeOH); 'HNMR reaction mixture was then diluted with 50 mL of EtOAc, and (DMs0-d~) 6 (doubling due to amide rotamers) 11.51 and 11.32 the DCU precipitate was filtered off. The filtrate was washed (s, lH), 8.86 and 8.68 (s, lH), 7.89-7.47 (m, 5H), 7.18 and 7.14 with HzO (4 x 50 mL). The organic was dried (NazS04) and (d, J = 8.2 Hz, 2H), 6.84 and 6.79 (d, J = 8.2 Hz, 2H), 5.16concentrated in uacuo to a thick oil. Chromatography (SiOz, 4.41 (m, 3H), 4.18 (s, 2H), 4.05 and 3.79 (d, J = 11.7, 4.1 Hz, 5% MeOWCHC13) provided 1.33 g (30%) of 8f as an amber

-

-

Triacid Angiotensin IZ Antagonists

Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26 4619

Table 3. Single-Crystal X-ray Crystallographic Analysis of 7h empirical formula Cz3H3oN406 formula weight 458.5 color; habit CLEARPABUM crystal system orthorhombic (Phosphonomethyl)phenoxy]-1-[ l-oxo-2(R)-[4[(%suIfobenzoyl)amino]-~-imidazol-l-yl]octyll-~-proline (le) and space group p212121 a = 9.0980(10)A unit cell dimensions cis-4-[4-(2-Phosphonoethyl)phenoxyl-l-[l-oxo-2(lt~-~4-~~2b = 10.694(2)A 8ulfobenzoyl)aminol-~ - i m i d a z o l - 1 - y l l ~ l l -(10. ~p~~e c = 25.388(6)A To a solution of 8e (4.70 g, 6.5 mmol) in 25 mL of anhydrous volume 2470.0(9)A3 CHzClz at 0 "C was added trimethylsilyl bromide (5.0 g, 32.4 z 4 mmol) dropwise over a 15 min period. The resulting mixture density (calcd) 1.233mg/m3 was warmed to room temperature and stirred for 1 h. The absorption coefficient 0.709 mm-I solvent was then removed in vacuo, and the residue was F(OO0) 976 dissolved in 16 mL of 2 N NaOH. After stirring for 1 h, the Siemens SHELXTL PLUS (VMS) system used reaction mixture was acidified to pH = 1.0 with 5 N HC1. The solution direct methods aqueous was extracted with 10% EtOWEtOAc (3 x 50 mL). refinement method full-matrix least-squares The organic was dried (Na2S04) and concentrated to give a temperature 22 "C solid residue that was dissolved in minimal absolute EtOH Cu Ka (1= 1.541 78 A) radiation and triturated with EtzO/hexanes. Isolation by filtration 20 range 0.0-116.0" provided 2.79 g (61%) of the phosphonic acid l e as a pale reflections collected 1968 -28.5" (c 1.0, MeOH); lH 1943 (Rint= 0.00%) yellow solid: mp 190°C decs [ a ] ~ independent reflections 1338 (R > 4.0dF)) observed reflections NMR (DMSO-&) 6 (doubling due to amide rotamers) 11.87 final R indicies (obs. data) R = 7.27%,R, = 8.29% and 11.72 (s, lH), 8.08 (s, lH), 7.84-7.44 (m, 5H), 7.11 (d, J = 8.2 Hz, 2H), 6.76 and 6.72 (d, J = 8.2 Hz, 2H), 5.42 and 5.26 -4.41 (m, 3H), 4.03 and 3.58 (d, J = 11.5, 2.1 Hz, lH), hook to a force displacement transducer. The bath chambers 3.75 and 3.45 (d, J = 11.5, 1.0 Hz, lH), 2.83 (d, J = 21.1 Hz, were maintained at 37 "C, aerated with 95% Od5% COZ,and 2H), 2.45-1.91 (m, 4H), 1.19-0.96 (m, 8H), 0.80 (m, 3H); FAB contained physiological solution of the following composition MS 693.5. Anal. ( C ~ O H ~ ~ N ~ OC, IIP H,S N. ) (mM): NaC1,117; glucose, 5.6; NaHzP04,l.O; MgS04,0.7; KC1, Data for If: isolated 26% yield by filtration of the acidified 5.2; CaClZ, 1.8;NaHC03,26; and phentolamine hydrochloride, reaction mixture; mp 230-234 "C dec; [ah -28.4" (c 1.1, 0.003. Rings were equilibrated for 1 h with 2 g of tension. MeOH); lH NMR (DMSO-de) 6 (doubling due t o amide rotaDuring the equilibration period, the tissues were washed by mers) 12.13 and 12.01 (s, lH), 8.74 (bs, lH), 7.86-7.46 (m, overflow every 15 min. Rings were then exposed to 10 nM 5H), 7.15 and 7.05 (d, J = 8.4 Hz, 2H), 6.78 and 6.74 (d, J = Ang I1 and were allowed t o contract until a steady state was 8.4 Hz, 2H), 5.41 and 5.22 (dd, J1 = JZ = 7.0 Hz, lH), 5.09 and reached. This challenge was repeated an hour later. Tissues 4.95 (m, lH), 5.07 and 4.46 (dd, J = 9.4, 1.0 Hz, lH), 4.01 and were then washed every 15 min for 1h. Cumulative concen3.67 (dd, J = 13.0, 4.5 Hz, lH), 3.94 and 3.46 (dd, J = 13.0, tration-response curves to Ang I1 (0.1 nM to 10pM) were then 1.0 Hz, lH), 2.68 (m, 2H), 2.55-1.96 (m, 4H), 1.71 (m, 2H), obtained. At the conclusion of the concentration-response 1.31-0.98 (m, 8H), 0.81 (m, 3H); FD MS 707. Anal. (C31H39curve, tissues were washed every 2 min until baseline tension N4011PS.l.BHCl) C, H, N. was reached and then every 15 min for 30 min. Ang I1 ci~-4-[4-(Carboxymethyl)phenoxy]-l-[ l-ox0-2(R)-[4-[(2antagonists were dissolved in DMSO, added in a 10 pL volume, ~ o b e ~ o y l ) a m i n o l - ~ - ~ ~ l - l - y l (41) l ~ ~ l land - ~allowed p ~ ~ eto incubate for 30 min before repeating the and cis-4-[3-(2-Carboxyethyl)phenoxyl-l-[l-oxo-2(R~-[4concentration response curve to Ang 11. Contractions to Ang [(2-sulfobenzoyl)amino]-1H-imidazol-1-ylloctyll- proI1 were expressed as a percent of the maximum contraction line (42). Compounds 41 and 42 were prepared exactly as obtained in the control curve (the first Ang I1 concentrationdescribed for la-d. response curve). ECSOS(concentration that contracted the Data for 41: isolated in 58% yield; mp 166-171 "C; [aln tissue to half the control maximum) for each were calculated -15.9" (c 1.0, MeOH); lH NMR (DMSO-ds)6 (doubling due t o using a four-parameter logistics model (NLIN, SAS Institute, amide rotamers) 12.14 and 12.07 ( 8 , lH), 8.86 and 8.84 (s, lH), Cary, NC). 7.86-7.46 (m, 5H), 7.24-7.12 (m, 2H), 6.97-6.84 (m, 2H), 5.43 Computationand Analysis Of KB. According t o Waud, a and 5.23 (dd, J1 = JZ= 7.0 Hz, lH), 5.14 and 4.97 (m, lH), reasonable function to model an empirical dose-response curve 5.11 and 4.45 (dd, J = 9.7,2.6 Hz, lH), 4.08 and 3.76 (dd, J = is the three-parameter logistic: 13.0, 3.9 Hz, lH), 3.85 and 3.52 (dd, J = 13.0, 1.3 Hz, lH), 3.43 and 3.40 (s, 2H), 2.59-2.0 (m, 4H), 1.31-0.96 (m, 8H), response = m d [ l (ED50(l/u))Sl (1) 0.81 (m, 3H). Anal. (C31H36N4010S)C, H, N. Data for 42: isolated in 48% yield; mp 153-167 "C; [ a l ~ where max = the maximum possible response, a = the agonist -27.9" (c 1.1, MeOH); lH NMR (DMSO-de)6 (doubling due to concentration, and s = steepness of the sigmoidal curve.28If a amide rotamers) 12.04 and 11.92 (s, lH), 8.56 (bs, lH), 7.86second dose response curve is generated after adding a 7.15 (m, 5H), 6.87-6.70 (m, 4H), 5.37-4.45 (m, 3H), 4.09competitive antagonist, then Waud (ref 30, eq 14) indicates 3.60 (m, 2H), 3.49 (s, 2H), 2.44-1.98 (m, 4H), 1.39-0.97 (m, the following equation relates equally effective agonist conH,SN. 8H), 0.81 (m, 3H). Anal. ( C ~ I H ~ ~ N ~ Oc,I O ) centrations: X-ray Crystallographic Analysis of 7h. X-ray crystallographic analysis of 7h was carried out by the X-ray Crystallography Department at Lilly Research Laboratories using a Siemens R3mN automated four-circle diffractometer. The structure was solved by direct methods using the Siemens where B = antagonist concentration, K g = dissociation conSHELXTL PLUS (VMS) system (Sheldrick, G. M. Shelxtl, stant of the antagonist, and A = agonist concentration equally Rev 4, Instrument Corporation, 1983). A summary of crystal effective in the presence of antagonist. Equation 2 may be parameters data collection and refinement is provided in Table substituted into eq 1 giving the following dose response 3. equation in the presence of a competitive antagonist: Pharmacological Methods. In Vitro Antagonism of Ang I1 (Rabbit Aorta). New Zealand white rabbits (Hazelresponse = m d [ l (ED50(l/A)(l (B/Kb))'I (3) ton, 2-3 kg) were sacrificed by cervical dislocation, and thoracic aortae were removed and cleaned of excess fat and If a second dose response curve is generated after adding a connective tissue. Rings of tissue (3 mm wide) were mounted noncompetitive antagonist, Kenakin suggests the following in 10 mL tissue baths between L-shaped stainless steel hooks. modification t o eq 2: The lower hook was attached to a stationary rod and the upper

lH), 3.81 and 3.45 (d, J = 11.7 Hz, 1.0 Hz, lH), 2.58-1.87 (m, 4H), 1.24-1.03 (m, 8H), 0.82 (m, 3H). Anal. (C31H36N808S) C, H, N. General Method for the Preparation of cis-444-

+

+

+

4520 Journal of Medicinal Chemistry, 1994, Vol. 37,No. 26

l/a = [(l/A)(l

+ (B/K,))+ intl

(4)

where int = intercept term for the linear equation.31 Equation 4 may be substituted into eq 1, giving the following dose response equation in the presence of a noncompetitive antagonist:

response = max/[l

+ (ED5&(1/A)(l+ (BK,)) t

For competitive antagonists, eqs 1and 3 were simultaneously fit t o pairs of dose response curves without and with antagonist, respectively. For noncompetitive antagonists, eqs 1and 5 were fit simultaneously. The curve fitting and estimation of KB were done by the nonlinear least squares methodology available in the software package JMP.32 The estimated KB values (after logarithmic transformation) were compared among compounds using analysis of variance with the TukeyKramer method for all painvise comparisons. In Vivo Antagonism of Ang I1 (Pithed Rats). Male Sprague-Dawley rats (Harlan Industries, Indianapolis, IN), 240-280 g, were anesthetized with isoflurane. The trachea was intubated with PE240 tubing and the rats pithed by insertion of a steel rod (1.5 mm diameter) down the spinal canal. Rats were immediately respired at a rate of 80 cycles/ min at a volume of 0.6 mu100 g of body weight. The right carotid artery and right jugular vein were cannulated for blood pressure monitoring and injection of drugs. Animals were allowed to equilibrate 15 min before a noncumulative iv dose response curve t o Ang I1 (10 ngkg to 10 pgikg) was obtained. For oral studies, rats were dosed by gavage 4 h prior to pithing. The in vivo KBfor each compound was calculated using either competitive or noncompetitive kinetic modeling as described above. Conscious SHR Monitored by Telemetry. Spontaneously hypertensive male rats [Tac:N(SHR)fBR]were obtained at 12-20 weeks of age from Taconic Farms (Germantown, NY) and housed under a 12 h diurnal cycle. Approximately 1week after arrival, rats were anesthetized with im ketamine (60 mg/ kg) followed in 5 min with ip pentobarbital (21 mgikg). Under sterile conditions, the abdomen was shaved and prepped with 2-propanol, and a 4.5 cm abdominal incision was made beginning just caudal t o the level of kidneys. The abdominal aorta was isolated and gently cleaned of connective tissue with a sterile cotton swab. A small spatula was used to raise a portion of the aorta away from the vena cava in an area just rostal to the iliac bifurcation. Two bulldog clamps isolated a portion of the aorta between the bifurcation and the left renal artery. The aorta was punctured near the iliac bifurcation clamp using a 21 G needle (bent at a 45” angle and with the bevel down). A fluid-filled catheter (0.7 mm OD, 8 cm in length) attached to a hermetically sealed sensor and radio transmitter (model TA11PA-C40, Data Sciences, St. Paul, MN) was inserted up to the rostal clamp using the bent needle as a guide. The needle was removed and the area dried with a cotton swab, and one drop of tissue adhesive was applied at the entry point while the clamps are removed. The entry point was then further sealed using tissue adhesive and a cellulose fiber patch (Vetbond, 3M Co.). The body of the transmitter was then sutured to the muscles of the inner abdominal wall using nonabsorbable 4-0 silk. The muscle layers were then approximated with interrupted knots using sterile 3-0 silk, and the final incision closed with sterile metal wound clips. All animals were administered 10 000 units of penicillin im (Ambi-Pen, Butler, Columbus, OH) and housed individually with food and water ad lib. All animals were allowed to recover for at least one week before study. Food access was restricted for the next 12 h while the animal’s blood pressure and heart rate were monitored by telemetry. At the time of study, rats were 13-50 weeks of age and weighed from 300 to 450 g. In some experiments, animals were briefly sedated with isoflurane (Aerrane, Anaquest, Madison, WI) and administered about 0.3 mL of Lasix (furosemide, 10 mgkg sc, Sigma, St. Louis, MO) in a

Palkowitz et al. vehicle consisting of 10% DMSO and PEG 200. Access to drinking water was restricted from the time of Lasix administration (about 5:OO P. M.) until the conclusion of the study. Twenty four hours after Lasix dosing, the rats were again briefly sedated with isoflurane and administered test compounds or vehicle by oral gavage (0.1 mL of 0.1 N NaOH per 30 mg of compound then diluted with distilled water so as t o administer a volume of approximately 2 mL). Pressure signals were acquired for 30 s every 10 min using a Data Sciencesprovided software package (Dataquest W ,version 2.0). The digitized values were stored and manipulated using a Compac Deskpro 486/33M computer and were corrected for ambient pressures. Six readings over a 1h period were averaged for analysis. Mean pressures for the 2 h period before dosing were taken as baseline value. Statistical analysis was performed using JMP software (SASInstitute, Cary, NC). A comparison of hourly postdrug mean pressure to baseline values within groups was performed using a paired sample Student’s t test. ANOVA was used for statistical comparisons among groups using the Tukey-Kramer HSD test for multiple comparisons. All painvise comparisons were made at each time interval up to 12 h.

Acknowledgment. The authors wish to thank the Physical Chemistry Department and X-ray Crystallography Laboratory at Lilly Research Laboratories for their collaboration in this work. SupplementaryMaterial Available: Atomic coordinates, bond lengths, bond angles, anisotropic displacement coefficients, and hydrogen coordinates for 7h (5 pages). Ordering information is given on any current masthead page.

References (1) Riegger, G. A. J. Lessons from recent randomized controlled

trials of the management of congestive heart failure. A m . J . Cardiol. 1993, 71, 38-403. (2) Rush, J. E.; Rajfer, 9. I. Theoretical basis for the use of angiotensin I1 antagonists in heart failure. J. Hypertens. 1993, 11, 569-71. (3) Black, M. J.; Campbell, J. H.; Campbell, G. R. Effect of perindopril on cardiovascular hypertrophy of the SHR: Respective roles of reduced blood pressure and reduced angiotensin I1 levels. Am. J . Cardiol. 1993, 71, 17-21E. (4) Lewis, E. J.; Hunsicker, L. G.; Bain, R. P.; Rohde, R. D. The effect of angiotensin-converting enzyme inhibition on diabetic nephropathy. N.Engl. J . Med. 1993,329, 1456-1462. ( 5 ) Hui, K. Y., Haber, E. Renin Inhibitors. In Robertson, J. I. S., Nicholls, M. G., Eds. The Renin Angiotensin System; London: Mosby, 1993; Vol. 2; pp 85.1-85.14. (6) Ocain, T. D.; Abou-Gharbia, M. Renin-angiotensin system (RAS) dependent antihypertensive agents: 1.Renin inhibitors. Drugs Future 1991, 16, 37-51. (7) Carini, D. J ; Duncia, J. V.; Aldrich, P. E.; Chiu, A. T.; Johnson, A. L.; Pierce, M. E.; Price, W. A.; Santella 111, J. B.; Wells, G. J.; Wexler, R. R.; Wong, P. C.; Yoo, S.-E.; Timmemans, P. B. M. W. M. Nonpeptide angiotensin I1 receptor antagonists: The discovery of a series of N-(biphenylmethy1)imidazolesas potent, orally active antihypertensives. J.Med. Chem. 1991,34,25252547. (8) For recent reviews see: (a) Steinberg, M. I.; Wiest, S. A.; Palkowitz, A. D. Nonpeptide angiotensin I1 antagonists. Cardiouasc. Drug. Reu. 1993,11, 312-358. (b) Timmermans, P. B. M. W. M.; Wong, P. C . ; Chiu, A. T.; Herblin, W. F.; Benfield, P.; Carini, D. J.;Lee, R. J.; Wexler, R. R.; Saye, J. M.; Smith, R. D. Angiotensin I1 receptors and angiotensin I1 receptor antagonists. Pharmacol. Rev. 1993,45, 205-251. (9) Okunishi, H.; Song, K.; Oka, Y.; Kobayashi, T.; Kawamoto, T.; Ishigara, H.; Mori, N.; Miyazaki, M. In vitro pharmacology of a novel nonpeptide angiotensin I1 receptor antagonist, E4177. Jpn. J. Pharmacol. 1993,62, 239-244. (10)Reitz, D. B.; Garland, D. J.; Norton, M. B.; Collins, J. T.; Reinhard, E. J.; Manning, R. E. N1-Sterically hindered 2Himidazol-2-one angiotensin I1 receptor antagonists: the conversion of surmountable antagonists to insurmountable antagonists. Bioorg. Med. Chem. Lett. 1993,3, 1055-1060. (11) Shibouta, Y.; Inada, Y.; Ojima, M.; Wada, T.; Noda, M.; Sanada, T.; Kubo, K.; Kohara, Y.; Naka, T.; Nishikawa, K. Pharmacological profile of a highly potent and long-acting angiotensin I1 receptor antagonist, 2-ethoxy-l-[[2’-(1H-tetrazol-5-y1)biphenyl-

Triacid Angiotensin ZZ Antagonists 4-yllmethyll-1H-benzimidazole-7-carboxylic acid (CV-11974),and its prodrug, (~)-l-(cyclohexyloxycarbonyloxy)-ethyl-2-ethoxy-l-

Journal of Medicinal Chemistry, 1994, Vol. 37, No. 26 4521

(21) McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna, M.-C. The facile dealkylation of phosphonic acid dialkyl esters by [~2’-~1H-tetrazol-5-yl~biphenyl-4-yllmethyll-1H-benzimidazole-7- bromotrimethylsilane. Tetrahedron Lett. 1977, 155-158. carboxylate (TCV-116). J . Pharmacol. Exp. Ther. 1993, 266, (22) Lin, H.-S.; Rampersaud A. A.; Zimmerman K.; Steinberg, M. I.; 114-120. Boyd D. B. Nonpeptide angiotensin I1 receptor antagonists: (12) Steinberg, M. I.; Palkowitz, A. D.; Thrasher, K. J.; Reel, J. K.; Synthetic and computational chemistry of N-[[4-[2-(2H-tetrazolZimmerman, K. M.; Whitesitt, C. A.; Simon, R. L.; Hauser, K. 5-yl)-l-cycloalken-l-yllphenyllmethyl]imidazole derivatives and L.; Lifer, S. L.; Pfeifer, W.; Takeuchi, IL;Wiest, S. A.; Vasudevan, their i n vitro activity. J . Med. Chem. 1992, 35, 2658-2667. V.; Bemis, K. G.; Deeter J. B.; Barnett, C. J.; Wilson, T. M.; (23) All compounds in Chart 2 were evaluated as racemic mixtures. Marshall, W. S.; Boyd, D. B. Chiral recognition of the angiotensin I1 (AT1)receptor by a highly potent phenoxyproline octanoamide. (24) Wong, P. C.; Price, W.; Chiu, A. T.; Duncia, J. V.; Carini, D. J.; Bioorg. Med Chem. Lett. 1994, 4, 51-56. Wexler, R. R.; Johnson, A. L.; Timmermans, P. B. M. W. M. (13) The compounds in Chart 2 were prepared as described in the Nonpeptide angiotensin I1 receptor antagonists. XI. Pharmacolfollowing: Lifer, S. L.; Marshall, W. S.; Mohamadi, F.; Reel, J. ogy of EXP3174: a n active metabolite of DuP 753, an orally K.; Simon, R. L.; Steinberg, M. I.; Whitesill, C. A. Eur. Pat. active antihypertensive agent. J . Pharmacol. Exp. Ther. 1990, 438869A, 1991. 255, 211-217. (14) Compounds 16-7, 19-40 were prepared as described for la(25) Losartan was prepared at Lilly Research Laboratories according d. Compound 18 was prepared in an analogous manner to l f , to the procedures described in ref 7. llheptane. See: Bowersemploying 5-aza-2-oxa-3-oxobicyclo[2.2. Nemia, M. M.; Joulli6, M. M. Heterocycles 1983, 20 (5), 817(26) Robertson, M. J.; Barnes, J. C.; Drew, G. M.; Clark, K. L.; 828. Marshall, F. H.; Michel, A.; Middlemiss, D.; Ross, B. C.; Scopes, (15) Krapcho, J.; Turk, C.; Cushman, D. W.; Powell, J. R.; DeForrest, D.; Dowle, M. D. Pharmacological profile of GR117289 i n vitro. J. M.; Spitzmiller, E. R.; Karanewsky, D. S.; Duggan, M.; A novel, potent and specific nonpeptide angiotensin AT1 receptor Rovnvak, G.; Schwartz, J.;Natarajan, S.; Godfrey, J. D.; Ryono, antagonist. Br. J. Pharmacol. 1992,107, 1173-1180. D. E.; Neubeck, R.; Atwal, K. S.;Petrillo, E. W., Jr. Angiotensin(27) For an examination of all seven diastereomers of l b , see ref 12. converting enzyme inhibitors. Mercaptan, carboxyalkyl dipep(28) Kauffman, R. F.; Bean J . S.; Zimmerman, K. M.; Brown, R. F.; tide, and phosphonic acid inhibitors incorporating 4-substituted Steinberg, M. I. Losartan, a nonpeptide angiotensin I1 (Ang 11) prolines. J . Med. Chem. 1988, 31, 1148-1160. (16) Mitsunobu, 0. The use of diethyl azodicarboxylate and triphreceptor antagonist inhibits neointima formation following balenylphosphine in synthesis and transformation of natural loon injury to rat carotid arteries. Life Sci. 1991,49, PL-222products. Synthesis 1981, 1-28. 228. (17) For 2g, see: Dillard, R. D.; Carr, F. P.; McCullogh, D.; Haisch, (29) Weinstock, J.; Keenan, R. M.; Samanen, J.; Hempel, J.; FinkelK. D.; Rinkema, L. E.; Fleisch, J. H. Leukotriene receptor stein, J. A.; Franz, R. G.; Gaitanopoulous, D. E.; Girard, G . R.; antagonists. 2. The [[(tetrazol-5-aryl)oxylmethyllacetophenone Gleason, J. G.; Hill, D. T.; Morgan, T. M.; Peishoff, C. E.; Aiyar, derivatives. J . Med. Chem. 1987,30, 911-918. N.; Brooks, D. P.; Frederickson, T. A.; Ohlstein, E. H.; Ruffolo, (18) For 2e, see: Ornstein, P. L.; Schaus, J. M.; Chambers, J. W.; R. R., Jr.; Stack, E. J.; Sulpizio, A. C.; Weidley, E. F.; Edwards, Huser, D. L.; Leander, J. D.; Wong, D. T.; Paschal, J. W.; Jones, R. M. l-(Carboxybenzyl)imidazole-5-acrylicAcids: Potent and N. D.; Deeter, J. B. Synthesis and pharmacology of a series of and -piperidine-2-carboxylic 3- and 4-(phosphonoalkyl)-pyridineselective angiotensin I1 receptor antagonists. J. Med. Chem. acids. Potent N-methyl-d-aspartate receptor antagonists. J . Med. 1991,34, 1514-1517. Chem. 1989,32,827-833. (30) Waud, D. R. InAdvances in General and Cellular Pharmacology; (19) For 2f, see: Hullar, T. L. Pyndoxyl phosphate. I. Phosphonic Narahashi, L. T., Bianchi, C. P., Eds.; Plenum: New York 1976; acid analogs of pyridoxyl phosphate. Synthesis via Wittig Vol. 1, Chapter 4, pp 145-178. reactions and enzymatic evaluation. J . Med. Chem. 1969,12, (31) Kenakin, T. P. The classification of drugs and drug receptors in 58-63. isolated tissues. Pharmacol. Rev. 1984, 36, 165-222. (20) The stereochemistry of 16 and 17 was assigned based on an (32) JMP User’s Guide: Version 2 of J M P SAS Institute, Inc.: Cary, independent synthesis from (E)-5 and L-proline methyl ester NC, 1989; Chapter 18, pp 427-450. (DCC coupling).